1013-00005 · Genome Dx Therapeutics Inc · Experts

expert-selection-qualification-memo.docx DOCUMENT Litigation

25 / 35 ← → ✕

MEMORANDUM

PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT

TO: Catherine Bloom, Responsible Partner, intellectual Property

FROM: Peter Vassiliev, Lead Associate, intellectual Property

DATE: January 2026

RE: Expert Selection and Qualification — Genome Dx Therapeutics Inc. v. Helios Precision Diagnostics, Inc., Case No. 3:25-cv-01847-JWL (N.D. Cal.)

CLIENT: Genome Dx Therapeutics Inc. ("GenomeDx" or "GDT")

MATTER: Patent infringement action asserting U.S. Patent Nos. 10,438,271 ("the '271 Patent"), 11,204,893 ("the '893 Patent"), and 11,891,647 ("the '647 Patent") against Helios Precision Diagnostics, Inc. ("Helios") and its ProstaSignal™ Genomic Classifier

I. Purpose and Executive Summary

This memorandum evaluates candidate technical and damages experts for plaintiff GenomeDx's affirmative patent infringement case against Helios Precision Diagnostics, Inc. and provides recommendations for the prompt retention of experts in both categories. The matter is progressing on an accelerated schedule, and early identification, vetting, and engagement of qualified experts is critical to meeting the deadlines established in the Court's Scheduling Order and to ensuring the highest quality of expert work product at the Markman hearing and at trial.

Two categories of experts are required: (1) a technical expert with deep knowledge of genomic classification methods who can opine on infringement analysis (including literal infringement and the doctrine of equivalents), patent validity in the face of Helios's invalidity contentions, and claim construction of the disputed claim terms; and (2) a damages expert with substantial experience in patent damages quantification who can present opinions on lost profits, reasonable royalty calculations, and a comprehensive Georgia-Pacific factor analysis tailored to the molecular diagnostics industry.

After evaluating multiple candidates in each category, I recommend the following:

• Technical Expert: Dr. Franklin Ogawa, PhD, Professor of Computational Oncology, Stanford University. Dr. Ogawa possesses precisely targeted expertise in machine-learning classifiers applied to genomic data, has an extensive publication record, and has testified as a technical expert in three prior patent infringement matters involving molecular diagnostics and bioinformatics — with his testimony admitted without challenge under Daubert in both the Northern District of California and the District of Delaware.

• Damages Expert: Dr. Louise Tremaine, PhD, Managing Director, Ashworth Economics Group LLC, Chicago, Illinois. Dr. Tremaine has served as a damages expert in more than twenty federal patent cases, has never had testimony excluded under Daubert, and has particular depth in life sciences, diagnostics, and medical device patent damages — including familiarity with Medicare reimbursement structures and the MolDX program, which are directly relevant to this matter.

The key case deadlines relevant to expert retention and preparation are as follows:

Draft engagement letters are being prepared concurrently for both recommended experts, and I anticipate that, subject to your approval, we can execute those letters and onboard both experts during February 2026 — providing adequate lead time for document review, consultation, and report preparation.

As background on the current posture of the case: the Court denied Helios's Motion to Dismiss under 35 U.S.C. § 101 on January 13, 2026, rejecting Helios's argument that the asserted claims are directed to patent-ineligible subject matter. The parties are currently in active fact discovery, which is scheduled to close on March 6, 2026. Claim construction briefing preparation is underway, with the Markman hearing set for September 18, 2026. The case remains on track for a February 2027 trial before Judge J. William Liang.

II. Technical Expert — Requirements and Evaluation Criteria

A. Scope of Technical Expert's Anticipated Opinions

The technical expert retained by GenomeDx will be expected to provide opinions across several interrelated areas. The scope of the anticipated opinions is summarized below and will be refined as fact discovery is completed and as the claim construction positions of both parties crystallize.

Infringement Analysis. The core of the technical expert's engagement will be an infringement analysis addressing whether Helios's ProstaSignal™ test infringes the asserted claims of the '271, '893, and '647 Patents. The ProstaSignal™ test, as currently understood from Helios's public disclosures and preliminary discovery responses, utilizes a 24-gene expression panel, processes gene expression data through a gradient-boosted decision tree classifier, and generates a ProstaScore ranging from 0 to 100 that places patients into Low, Intermediate, or High risk classifications. The expert will need to map each element of the asserted claims onto the corresponding features of the ProstaSignal™ system, with reference to both the patent specifications and the prosecution histories of each asserted patent.

Literal Infringement and Doctrine of Equivalents. The technical expert must be prepared to opine on both literal infringement and infringement under the doctrine of equivalents. The doctrine of equivalents analysis is particularly important with respect to the '893 Patent, which recites a "random-forest classifier" limitation in several of its asserted claims. Helios's ProstaSignal™ test employs a gradient-boosted decision tree classifier rather than a random-forest classifier. The expert will need to analyze whether Helios's gradient-boosted decision tree classifier performs substantially the same function, in substantially the same way, to achieve substantially the same result as the claimed random-forest classifier — or, alternatively, whether the two classifiers are insubstantially different under the Harborline Enterprises Davis formulation. This analysis will require deep technical understanding of both ensemble learning methods and the specific algorithmic and functional differences (and similarities) between random forests and gradient-boosted decision trees as applied to gene expression data.

Gene Overlap Analysis. The technical expert should also address the significance of the 19-of-24 gene overlap between ProstaSignal™'s 24-gene panel and GenomeDx's patented gene set. This overlap is relevant to multiple claim limitations across the three asserted patents, including the '271 Patent's claim limitation requiring "a panel of at least 22 genes selected from a predefined gene set." The expert will need to consider the degree of overlap, the functional significance of the shared genes, and the implications of the remaining genes in the ProstaSignal™ panel that fall outside the patented gene set.

Validity Opinions. The technical expert must be prepared to respond to Helios's invalidity contentions under 35 U.S.C. §§ 101, 102, and 103. Helios's preliminary invalidity contentions identify the following principal prior art references:

• Nakamura et al., U.S. Patent No. 9,617,340 — directed to gene expression-based classification methods for cancer prognosis, but employing a different classifier architecture and training methodology.

• Gagnon, WO 2018/065413 — an international patent publication describing multi-gene panels for urological cancer risk assessment.

• Thornberry et al., "Machine Learning Approaches to Genomic Risk Stratification in Prostate Cancer," Journal of Computational Oncology, Vol. 14, No. 3, pp. 287–301 (March 2015) — an academic publication describing early-stage research on machine-learning classifiers applied to prostate cancer gene expression data.

• Klein et al., "A 17-Gene Assay to Predict Prostate Cancer Aggressiveness in the Context of Gleason Grade Heterogeneity, Tumor Multifocality, and Biopsy Undersampling," European Urology, Vol. 66, No. 1, pp. 45–52 (2014) — describing a commercially available multi-gene assay for prostate cancer risk assessment.

The expert must be able to distinguish each of these references from the asserted claims and, where necessary, demonstrate that the claimed inventions would not have been obvious to a person of ordinary skill in the art at the time of the inventions in view of these references, whether taken individually or in combination.

Claim Construction Testimony. The technical expert may be called upon to provide testimony at the Markman hearing scheduled for September 18, 2026. The parties' joint claim construction statement identifies the following disputed claim terms, among others:

• "trained machine-learning classifier"

• "random-forest classifier"

• "panel of at least 22 genes selected from a predefined gene set"

• "risk category"

• "composite risk metric"

The expert should be able to explain to the Court how a person of ordinary skill in the art at the relevant time would understand each of these terms in the context of the patent specifications, the prosecution histories, and the state of the art in computational oncology and genomic diagnostics.

Section 101 / Patent Eligibility. Although the Court has already denied Helios's Motion to Dismiss under § 101, it is possible that Helios will reassert patent-eligibility arguments at the claim construction stage or at summary judgment. The expert must be prepared to address Helios's contention that the asserted claims are directed to abstract mathematical algorithms applied to natural phenomena (i.e., gene expression levels), citing Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012), and Association for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013). The expert must be able to articulate, from a technical perspective, the specific, concrete diagnostic improvements recited in the claims — including the particular combination of sample processing steps, gene selection, classifier training methodology, and clinically actionable risk categorization — that distinguish the claimed inventions from mere abstract ideas or laws of nature.

B. Desired Qualifications

In selecting a technical expert, we evaluated candidates against the following criteria, which reflect both the substantive requirements of the case and the practical demands of persuasive expert testimony:

1. Academic Credentials. PhD-level training in computational biology, bioinformatics, computational oncology, or a closely related field. An advanced degree from a recognized research institution is a baseline requirement, and active affiliation with an academic institution or major research center is strongly preferred.

2. Subject-Matter Expertise. Deep expertise in machine-learning classifiers applied to genomic data, with particular knowledge of random forests, gradient-boosted decision trees, and other ensemble methods. The candidate must be conversant with the algorithmic principles underlying these classifiers, their application to high-dimensional gene expression data, and the practical and theoretical differences between the various approaches.

3. Familiarity with Laboratory Methodologies. Working knowledge of gene expression profiling methodologies relevant to the asserted patents, including processing of formalin-fixed, paraffin-embedded (FFPE) tissue specimens, RNA extraction protocols, reference-gene normalization techniques, and the technical challenges associated with deriving reliable gene expression measurements from archived clinical specimens.

4. Clinical Context. Understanding of prostate cancer molecular diagnostics and clinical genomic risk stratification. The expert should be able to situate the technical discussion within the clinical context, explaining why and how genomic classifiers are used to inform treatment decisions for prostate cancer patients.

5. Publication Record. An active and substantial record of peer-reviewed publications in relevant journals, demonstrating recognition by the scientific community and ongoing engagement with the field.

6. Litigation Experience. Prior experience testifying as an expert witness in patent cases, preferably in the life sciences or molecular diagnostics domain. The candidate must be able to communicate complex technical concepts clearly to a judge and jury, and must have a track record that demonstrates the ability to withstand cross-examination and Daubert challenges.

7. Absence of Conflicts. No conflicts of interest with Helios, its investors (including Corinth Ventures Fund III, LP), or its outside litigation counsel (Ridgecrest Durneval LLP). Any prior relationships with parties, counsel, or key witnesses must be evaluated for potential appearance issues.

C. Candidate 1 (Recommended): Dr. Franklin Ogawa, PhD

1. Background and Qualifications. Dr. Franklin Ogawa is a Professor of Computational Oncology at Stanford University, where he has held a faculty appointment for more than fifteen years. He received his PhD in computational biology from the Massachusetts Institute of Technology and completed post-doctoral training at the Broad Institute of MIT and Harvard, where his research focused on machine-learning applications to cancer genomics — specifically, the development and validation of ensemble classifier methods for gene expression-based prognosis.

Dr. Ogawa has accumulated over twenty years of academic experience working at the intersection of bioinformatics, machine learning, and cancer diagnostics. His laboratory at Stanford is one of the leading groups in the country studying the application of computational methods to molecular oncology, and his work has directly contributed to the development and refinement of the analytical frameworks that underpin the technologies at issue in this case.

Dr. Ogawa has published more than eighty peer-reviewed articles in leading journals, including Nature Methods, Bioinformatics, the Journal of Computational Oncology, Clinical Cancer Research, European Urology, and Genome Research. His publications span a broad range of topics directly relevant to this matter, including genomic classifiers for cancer risk stratification, random forest and gradient-boosted decision tree models applied to gene expression data, and molecular risk stratification in urological cancers. Several of his publications specifically address the relative performance characteristics of random forest and gradient-boosted tree classifiers in genomic applications — a central technical issue in the '893 Patent infringement analysis.

Dr. Ogawa serves on the editorial boards of several relevant journals and is a frequent invited speaker at computational oncology conferences, including the annual meetings of the American Association for Cancer Research (AACR) and the International Society for Computational Biology (ISCB). He has served as a peer reviewer for National Institutes of Health (NIH) grants in the area of genomic diagnostics, further attesting to his standing in the field and his ability to evaluate the technical merit of diagnostic innovations.

2. Prior Testimony and Litigation Experience. Dr. Ogawa has served as a technical expert in three prior patent infringement matters involving molecular diagnostics and bioinformatics methodologies. In each of those engagements, he provided opinions on infringement, validity, and/or claim construction, and prepared detailed expert reports and tutorials for the court.

His testimony has been admitted without challenge under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), in the Northern District of California (before Judge Chen) and in the District of Delaware (before Judge Andrews). Opposing counsel in those matters did not file motions to exclude his testimony, and in one matter, the court specifically noted the thoroughness and clarity of Dr. Ogawa's expert report in its claim construction opinion.

Dr. Ogawa is experienced in preparing claim construction tutorials for Markman hearings, infringement analyses presented in claim-chart format, and validity opinions addressing both anticipation and obviousness. He has prepared materials for both bench trials and jury trials and has demonstrated an ability to explain complex technical concepts in accessible, persuasive terms.

3. Relevance to This Matter. Dr. Ogawa's qualifications are exceptionally well-suited to the technical issues presented in this case. His relevance to the specific matters at issue is summarized as follows:

First, Dr. Ogawa possesses specific expertise in the precise technical area at issue: machine-learning classifiers, including random forests and gradient-boosted decision trees, applied to genomic data derived from formalin-fixed, paraffin-embedded (FFPE) prostate cancer specimens. This is not merely adjacent expertise — it is the core of his research program and publication record.

Second, Dr. Ogawa can credibly opine on whether gradient-boosted decision tree classifiers are equivalent to random-forest classifiers under the doctrine of equivalents, which is the pivotal technical question in the '893 Patent infringement analysis. His publications comparing the two classifier families in genomic applications will provide a strong foundation for his opinions, and his academic standing will lend weight to his conclusions.

Third, Dr. Ogawa is well-positioned to address the technical significance of the 19-of-24 gene overlap between ProstaSignal™'s gene panel and GenomeDx's patented gene set. He can explain the biological and computational rationale for gene selection in panel-based classifiers and can opine on the relationship between ProstaSignal™'s 24-gene panel and the claim limitation requiring "a panel of at least 22 genes selected from a predefined gene set."

Fourth, Dr. Ogawa is well-positioned to rebut Helios's § 101 arguments by explaining, from a technical perspective, why the claimed methods represent specific, concrete improvements in diagnostic methodology — rather than abstract mathematical algorithms or laws of nature. His understanding of both the computational and the clinical dimensions of genomic classification will enable him to articulate the inventive concept embodied in the claims in terms that are both technically rigorous and accessible to the Court.

Fifth, Dr. Ogawa can address the prosecution history of the asserted patents, including: (a) the amendment made during prosecution of the '271 Patent that added specific technical steps (RNA extraction from FFPE samples and reference-gene normalization) to overcome prior art rejections; and (b) the distinction drawn during prosecution of the '893 Patent between the Nakamura reference and the claimed random-forest classifier, which was based on the training methodology and minimum cohort size requirements specified in the claims.

4. Conflict Check. We have conducted a preliminary conflict check on Dr. Ogawa, with the following results:

• Dr. Ogawa has no prior consulting or employment relationship with Helios Precision Diagnostics, Inc., Corinth Ventures Fund III, LP (Helios's lead investor), Ridgecrest Durneval LLP (Helios's outside litigation counsel), or any scientists currently or formerly affiliated with Helios.

• Dr. Ogawa has no disclosed financial interest in GenomeDx, Pinnacle Biosciences Capital, LP (GenomeDx's primary institutional investor), or Ridgeline Health Ventures.

• Dr. Ogawa has no prior professional relationship with Dr. Elaine Kossack, Helios's Chief Scientific Officer and Broad Institute alumna, or with Dr. Nathan Ashford, Helios's Chief Executive Officer, that would create an appearance of bias or partiality.

• Dr. Ogawa has co-authored one publication with Dr. Rajan Mehta, GenomeDx's Chief Technology Officer and a named inventor on all three asserted patents. The co-authored publication was a 2018 article in Bioinformatics on the application of deep learning architectures to gene expression data, approximately eight years ago. We have assessed whether this prior co-authorship could be used by the defense to challenge Dr. Ogawa's independence and conclude that a single co-authored paper from 2018, on a topic tangentially related to but distinct from the specific inventions at issue, is unlikely to create a disqualifying conflict or to provide meaningful ammunition on cross-examination. However, we recommend that this relationship be disclosed preemptively in Dr. Ogawa's expert report and, if necessary, addressed directly during his testimony to neutralize any potential line of attack.

5. Daubert Considerations. Dr. Ogawa presents an exceptionally strong profile for Daubert admissibility. His qualifications are squarely within the technical subject matter at issue, and there is no gap between his area of expertise and the opinions he will be asked to render. His methodology will be based on well-accepted principles of computational biology and machine learning, including standard analytical frameworks for comparing classifier architectures, evaluating gene panel composition, and assessing the technical significance of algorithmic choices in genomic diagnostics.

His prior judicial acceptance — specifically, the admission of his testimony without challenge in two federal districts — provides additional support for admissibility in this matter. His extensive publication record in peer-reviewed journals demonstrates that his views are recognized and credited by the relevant scientific community, further bolstering the reliability prong of the Daubert analysis.

We do not anticipate a significant risk of exclusion of Dr. Ogawa's testimony, although we will, of course, prepare a thorough Daubert defense brief should Helios move to exclude.

D. Alternative Technical Expert Candidates Considered

In addition to Dr. Ogawa, we evaluated two alternative candidates for the technical expert role. Both were ultimately not selected for the reasons described below.

1. Candidate 2: Dr. Marcia Ellison, PhD. Dr. Marcia Ellison is an Associate Professor of Bioinformatics at the University of Michigan. She holds a PhD in bioinformatics and has published approximately thirty-five peer-reviewed articles on gene expression analysis, with a focus on transcriptomic profiling in solid tumors. Her academic credentials are solid and her publication record is respectable within her subfield.

However, Dr. Ellison's qualifications are narrower than Dr. Ogawa's in several respects that are directly relevant to this matter. Most significantly, Dr. Ellison lacks specific experience with ensemble classifiers — she has not published on random forests or gradient-boosted decision trees in the context of genomic data, and her machine-learning expertise is concentrated in other methodological areas (principally neural networks and support vector machines). This gap would make it more difficult for her to provide authoritative opinions on the doctrine-of-equivalents analysis for the '893 Patent, which requires a nuanced comparison of random forests and gradient-boosted decision trees.

Additionally, Dr. Ellison has never previously testified as an expert witness in patent litigation or in any other federal proceeding. While a lack of prior testimony is not disqualifying per se, it introduces a higher degree of Daubert risk and uncertainty regarding her performance at deposition and at trial. First-time expert witnesses may be less effective in communicating with judges and jurors, and opposing counsel may attempt to exploit the absence of a track record.

No conflicts of interest were identified with respect to Dr. Ellison.

Recommendation: Not selected. Dr. Ellison's less directly relevant expertise, particularly her lack of experience with ensemble classifiers, and her absence of prior testimony experience create unnecessary risk. Dr. Ogawa is the superior candidate.

2. Candidate 3: Dr. Tobias Engström, PhD. Dr. Tobias Engström is an independent consultant and former Director of Bioinformatics at a major academic medical center. He holds a PhD in computational genomics and has a strong publication record of approximately fifty papers in genomic diagnostics, including several papers on ensemble methods in gene expression analysis. His technical qualifications are substantial, and he has testified as an expert witness on two prior occasions in molecular diagnostics patent cases — both times as a defense expert.

However, a potential issue exists with respect to Dr. Engström: he previously served as a testifying expert for Ridgecrest Durneval LLP, Helios's outside litigation counsel in this matter, in an unrelated patent infringement case approximately four years ago. While this prior engagement does not constitute a formal conflict of interest — the matters are unrelated and Dr. Engström's retention by Ridgecrest Durneval was on behalf of a different client — the relationship creates a perception of divided loyalty that could be exploited on cross-examination. Helios's counsel could argue that Dr. Engström is a "hired gun" willing to testify for either side, or could use the prior relationship to suggest that Dr. Engström has personal knowledge of Ridgecrest Durneval's litigation strategies, creating an uncomfortable dynamic.

Additionally, the fact that Dr. Engström's prior two testimonial engagements were both as a defense expert could create a narrative risk — opposing counsel might argue that Dr. Engström has now "switched sides" and is willing to offer whatever opinion his retaining counsel requests.

No other conflicts were identified.

Recommendation: Not selected. The prior relationship with defendant's counsel creates unnecessary risk, even though no formal conflict exists. The potential for cross-examination on this issue, and the perception of divided loyalty, weigh against retention. Dr. Ogawa is the superior candidate.

III. Damages Expert — Requirements and Evaluation Criteria

A. Scope of Damages Expert's Anticipated Opinions

The damages expert retained by GenomeDx will be expected to provide comprehensive opinions quantifying the damages suffered by GenomeDx as a result of Helios's infringement of the asserted patents. The anticipated scope of the damages expert's opinions encompasses the following areas:

Lost-Profits Analysis. GenomeDx's primary damages theory is based on lost profits. GenomeDx is the market leader in genomic prostate cancer prognostic testing through its Decipher® product. Prior to the launch of Helios's ProstaSignal™ test in approximately October 2024, GenomeDx held approximately 74% of the relevant market. Following the launch of ProstaSignal™, GenomeDx's market share declined to approximately 68%, representing a material erosion attributable to Helios's infringing product.

The damages expert will be expected to calculate lost profits based on the following framework, which will be refined as discovery progresses:

• Helios's estimated ProstaSignal™ test volume from October 2024 through March 2025: approximately 8,400 tests.

• Estimated number of tests diverted from GenomeDx: approximately 6,100 tests (based on the two-supplier market analysis and demand transfer theory described below).

• GenomeDx's average reimbursement per test (Decipher®): $4,210.

• Lost revenue attributable to diverted tests: approximately $25,681,000 (6,100 tests × $4,210 per test).

• GenomeDx's incremental margin on additional test volume: approximately 72%.

• Estimated lost profits: approximately $18,490,320 ($25,681,000 × 72%).

These figures are preliminary and will be updated as additional financial data is produced by both parties during the remaining weeks of fact discovery.

Two-Supplier Market Theory. To support the lost-profits claim, the expert will need to establish that GenomeDx and Helios are functionally in a two-supplier market for genomic prostate cancer prognostic testing. This analysis will require examination of the market for genomic tests that provide a clinically actionable risk classification for post-surgical prostate cancer patients, the absence of non-infringing substitutes that perform the same clinical function, and the barriers to entry in this market segment. The expert should be prepared to address the Panduit factors and to demonstrate GenomeDx's entitlement to lost profits under the established framework.

Reasonable Royalty as Alternative Damages Theory. In the event the Court determines that lost profits are not available or are not fully supported by the evidence, the damages expert will also present a reasonable royalty analysis as an alternative (or floor) damages theory. The preliminary reasonable royalty calculation is as follows:

• Royalty base: $32,088,000 (Helios ProstaSignal™ revenue from October 2024 through March 2025, calculated as 8,400 tests × $3,820 average reimbursement per test).

• Proposed royalty rate: 18%.

• Reasonable royalty: $5,775,840 ($32,088,000 × 18%).

Georgia-Pacific Factor Analysis. The reasonable royalty rate will be supported by a comprehensive analysis across all fifteen Georgia-Pacific factors, as set forth in Georgia-Pacific Corp. v. United States Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970), modified and affirmed, 446 F.2d 295 (2d Cir. 1971). Particular focus will be placed on Factor 12 (the portion of the profit or selling price that may be customary in the particular business or in comparable businesses to allow for use of the invention or analogous inventions), which provides the strongest anchor for the proposed royalty rate.

The most relevant comparable license is the license agreement between GenomeDx and Veridian Molecular Labs, dated April 3, 2022. The Veridian license has the following key terms:

• Running royalty of 15% on net revenues.

• Minimum annual royalty of $500,000.

• Licensed field limited to bladder cancer diagnostics under the '271 and '893 Patents only (the '647 Patent was not licensed to Veridian).

The damages expert will analyze the comparability of the Veridian license to the hypothetical negotiation between GenomeDx and Helios and will propose an upward adjustment of the Veridian 15% rate to 18% to account for the following distinguishing factors:

(a) Litigation context. The hypothetical negotiation is deemed to occur against the backdrop of actual infringement, whereas the Veridian license was negotiated in a non-litigation context. The Federal Circuit has recognized that a patentee may command a higher royalty rate in a litigation-context hypothetical negotiation than in a voluntary licensing transaction.

(b) Direct competitive overlap. The Veridian license was limited to bladder cancer diagnostics — a different clinical application that did not place Veridian in direct competition with GenomeDx's core prostate cancer testing business. Helios's ProstaSignal™, by contrast, directly competes with GenomeDx's Decipher® test in the prostate cancer prognostic testing market, and the resulting competitive harm justifies a premium over the Veridian royalty rate.

(c) Additional patent coverage. The Veridian license covered only the '271 and '893 Patents. The hypothetical license to Helios would encompass all three asserted patents, including the '647 Patent, which provides additional claim coverage for the genomic classification technology at issue. The inclusion of an additional patent in the hypothetical license justifies a higher royalty rate.

Enhanced Damages. The damages expert should also be prepared to address the availability of enhanced damages under 35 U.S.C. § 284. While the legal standard for enhancement is determined by the Court, the damages expert's analysis of willfulness-related financial factors — including Helios's knowledge of GenomeDx's patents prior to launching ProstaSignal™, Helios's failure to design around the asserted claims, and the degree of competitive harm to GenomeDx — may inform the Court's enhancement analysis.

Ongoing Royalty / Permanent Injunction Analysis. In the event that GenomeDx seeks a permanent injunction or, in the alternative, an ongoing royalty for post-judgment infringement, the damages expert should be prepared to address the appropriate ongoing royalty rate, taking into account the adjudication of infringement and the changed dynamics of a post-judgment negotiation.

B. Desired Qualifications

The following qualifications were used to evaluate candidate damages experts:

1. Academic Credentials. An advanced degree (PhD preferred) in economics, finance, or a closely related quantitative field.

2. Patent Damages Expertise. Extensive experience in patent damages analysis, with particular depth in the life sciences and diagnostics sector. The candidate must be thoroughly familiar with the Georgia-Pacific framework, the Panduit factors for lost-profits analysis, comparable-license analysis, apportionment principles, and other methodologies commonly employed in patent damages cases.

3. Industry Knowledge. Familiarity with the economics of the molecular diagnostics industry, including Medicare reimbursement structures (particularly the MolDX program and MAAA codes), laboratory-developed test (LDT) market dynamics, and the competitive landscape for genomic testing products.

4. Publication and Speaking Record. A strong record of publications and conference presentations in the area of patent damages and intellectual property economics, demonstrating thought leadership and peer recognition.

5. Litigation Experience. Prior testimony as a damages expert in federal patent cases, with a demonstrable track record of Daubert admissibility and persuasive performance at trial.

6. Absence of Conflicts. No conflicts of interest with Helios, Corinth Ventures Fund III, LP, or Ridgecrest Durneval LLP.

C. Candidate 1 (Recommended): Dr. Louise Tremaine, PhD

1. Background and Qualifications. Dr. Louise Tremaine is a Managing Director at Ashworth Economics Group LLC, a leading economics consulting firm based in Chicago, Illinois, that specializes in intellectual property litigation support and damages analysis. Dr. Tremaine received her PhD in economics from the University of Chicago, with a specialization in industrial organization and intellectual property valuation.

Dr. Tremaine has accumulated over fifteen years of experience as a damages expert in patent litigation, during which time she has developed particular depth in life sciences, diagnostics, and medical device patent damages. Her practice encompasses the full range of patent damages methodologies, including lost-profits analysis, reasonable royalty calculations, comparable-license analysis, apportionment, and enhanced damages assessments.

Dr. Tremaine has published articles on reasonable royalty methodologies, apportionment, and lost-profits analysis in peer-reviewed economics journals and IP law journals, including the Journal of the Patent and Trademark Office Society and Economics of Innovation and New Technology. She is a regular speaker at conferences hosted by the American intellectual Property Law Association (AIPLA) and the Licensing Executives Society (LES), where she has presented on patent damages topics including reasonable royalty determination, the evolving standards for apportionment, and the economic analysis of FRAND licensing obligations.

2. Prior Testimony and Litigation Experience. Dr. Tremaine has testified as a damages expert in more than twenty federal patent cases across a wide range of technologies and industries, with a concentration in life sciences, diagnostics, and medical devices. Her testimony has been consistently admitted under Daubert, and she has never had testimony or an expert report excluded by a court.

Notably, Dr. Tremaine has served as a damages expert for both plaintiffs and defendants in patent cases, demonstrating objectivity and analytical rigor that is not tethered to any particular advocacy position. This balanced track record enhances her credibility and makes her less vulnerable to characterization as a "plaintiff's expert" or a "defendant's expert" on cross-examination.

Of particular relevance to this matter, Dr. Tremaine has specific experience with diagnostics industry economics, including Medicare reimbursement structures under the MolDX program administered by Palmetto GBA, the assignment and use of Molecular Analysis for Therapy Choice (MAAA) procedure codes, and the market dynamics of laboratory-developed tests (LDTs). This experience is directly relevant to analyzing GenomeDx's Decipher® test (MAAA code 0047M) and Helios's ProstaSignal™ test (MAAA code 0239M), including the reimbursement rates, payer mix, and revenue recognition patterns for each product.

3. Relevance to This Matter. Dr. Tremaine is well-positioned to provide credible and persuasive damages opinions in this matter for several reasons.

First, she can credibly quantify both lost-profits and reasonable-royalty damages using GenomeDx's financial data. GenomeDx's fiscal year 2024 revenue was $387.6 million, with EBITDA of $54.3 million (a 14.0% EBITDA margin) and an average reimbursement of $4,210 per Decipher® test. Dr. Tremaine's experience with diagnostics industry economics will enable her to work efficiently with this data and to construct a damages model that accurately reflects GenomeDx's cost structure, margin profile, and incremental economics.

Second, Dr. Tremaine has extensive experience analyzing comparable licenses in patent damages cases — a critical skill for this matter, where the Veridian Molecular Labs license is the primary comparable. She is well-suited to address the comparability of the Veridian license to the hypothetical negotiation at issue and to articulate the economic rationale for the upward adjustment from 15% to 18%.

Third, Dr. Tremaine can opine on market share dynamics, including GenomeDx's decline from approximately 74% to approximately 68% market share following the launch of ProstaSignal™, and can construct a rigorous two-supplier market analysis using accepted economic frameworks.

Fourth, her familiarity with competitor-entry scenarios in molecular diagnostics markets — derived from her prior engagements in similar cases — will enable her to address the demand-transfer analysis (estimating the number of tests diverted from GenomeDx to Helios) with credibility and precision.

4. Conflict Check. We have conducted a thorough conflict check on Dr. Tremaine and Ashworth Economics Group LLC, with the following results:

• Dr. Tremaine has no prior engagement by or on behalf of Helios Precision Diagnostics, Inc., Corinth Ventures Fund III, LP, Aether Life Sciences Partners, or Ridgecrest Durneval LLP.

• Dr. Tremaine has no financial interest in GenomeDx, Pinnacle Biosciences Capital, LP, or any affiliated entities.

• Ashworth Economics Group LLC has not previously been engaged by any party adverse to GenomeDx in any matter.

• No conflicts of interest have been identified.

5. Daubert Considerations. Dr. Tremaine presents the strongest possible Daubert profile among the candidates evaluated. Her extensive track record — more than twenty federal patent damages engagements, with no instances of exclusion — provides powerful evidence that her methodologies are reliable and that her qualifications are sufficient to support the opinions she will be asked to render.

The methodologies she employs — including the Georgia-Pacific framework, the Panduit lost-profits test, and comparable-license analysis — are well-established and routinely accepted by federal courts. Her academic credentials (PhD in economics from the University of Chicago) and her publication record in peer-reviewed economics and IP law journals further support the reliability of her opinions under the Daubert standard.

We do not anticipate a significant risk of exclusion of Dr. Tremaine's testimony.

D. Alternative Damages Expert Candidates Considered

In addition to Dr. Tremaine, we evaluated two alternative candidates for the damages expert role. Both were ultimately not selected for the reasons described below.

1. Candidate 2: Dr. Richard Halstead, PhD. Dr. Richard Halstead is a Principal at Centennial Valuation Partners LLC in New York, New York. He holds a PhD in finance and has strong quantitative credentials, with particular expertise in financial modeling and valuation. Dr. Halstead has served as a patent damages expert in approximately ten prior engagements.

However, the substantial majority of Dr. Halstead's patent damages experience has been in the semiconductor and software industries. He has limited experience in the life sciences and diagnostics sectors and has not previously been retained in a case involving genomic testing, Medicare reimbursement, or MolDX program economics. Retaining Dr. Halstead would require significant investment of time and resources to educate him on the genomic testing market, Medicare reimbursement structures, MAAA codes, and the specific competitive dynamics of the prostate cancer diagnostics market.

This knowledge gap creates two related risks. First, there is a heightened risk that Helios could challenge Dr. Halstead's qualifications under Daubert on the grounds that he lacks sufficient knowledge of the industry to render reliable opinions. While this challenge might not ultimately succeed, it would consume litigation resources and could create unnecessary uncertainty. Second, the steeper learning curve would increase the overall cost of expert engagement and the time required to prepare the expert report, compressing an already tight timeline.

No conflicts of interest were identified with respect to Dr. Halstead.

Recommendation: Not selected. Dr. Halstead's lack of life sciences industry experience creates risk that his opinions may not withstand Daubert challenge or may be less persuasive to the jury, and the steeper learning curve increases costs and timeline pressure.

2. Candidate 3: Dr. Karen Nussbaum, PhD. Dr. Karen Nussbaum is an independent economics consultant based in Washington, DC, with extensive experience in life sciences patent damages. She has served as a damages expert in approximately fifteen prior matters, many of which involved pharmaceutical, biotechnology, or diagnostics patents. Her qualifications and industry experience are comparable to Dr. Tremaine's, and she would be a strong candidate in the abstract.

However, Dr. Nussbaum is currently retained as a damages expert in two other active patent cases with trial dates in 2026. Given the volume of work required for expert report preparation, deposition, and trial testimony in those concurrent matters, there are serious concerns about Dr. Nussbaum's availability for the May 15, 2026 expert report deadline and the February 2027 trial in this case. Dr. Nussbaum's office has indicated that she could commit to the engagement, but has acknowledged that scheduling would be "challenging" and that some work might need to be delegated to other professionals at her firm — a less-than-reassuring response given the complexity and financial significance of this matter.

No conflicts of interest were identified with respect to Dr. Nussbaum.

Recommendation: Not selected. Scheduling constraints create an unacceptable risk of delay or inadequate preparation. The expert report deadline is firm, and GenomeDx cannot afford a damages expert who is overextended across multiple matters. Dr. Tremaine is the superior candidate.

IV. Engagement Structure and Logistics

A. Retention Terms

Both recommended experts — Dr. Franklin Ogawa (technical) and Dr. Louise Tremaine (damages) — will be retained through Calderwood & Harkness LLP on behalf of GenomeDx. Standard expert engagement letters are being drafted for each expert and will be circulated for review promptly upon approval of the expert selections. Separate draft engagement letters are being prepared for each expert and will be provided under separate cover.

Both experts will be subject to the Stipulated Protective Order entered by Judge Liang on December 18, 2025, including the "Highly Confidential — Attorneys' Eyes Only" designation tier, which will govern their access to ProstaSignal™ algorithm details, Helios's financial data, and GenomeDx's proprietary information. Each engagement letter will include an acknowledgment by the expert of the Protective Order's terms, and each expert will execute the Protective Order's required undertaking prior to receiving any confidential or highly confidential materials.

Hourly rates for each expert will be negotiated during the engagement letter process. Based on preliminary discussions, we anticipate that Dr. Ogawa's rate will be in the range of $800–$950 per hour and Dr. Tremaine's rate will be in the range of $900–$1,100 per hour, consistent with prevailing market rates for experts of their caliber in the Northern District of California.

B. Timeline and Deliverables

The retention target for both experts is February 2026, with engagement letters to be executed promptly following client approval and final conflict confirmations. The following table summarizes the key milestones and the expert-related deliverables associated with each:

Given the expert report deadline of May 15, 2026, and the volume of material that both experts will need to review and analyze, it is critical that we onboard the experts as early as possible in February 2026 and begin providing them with initial document packages and case background materials without delay.

C. Document Access and Coordination

Both experts will receive access to the Canfield Document Solutions LLC "RevealIQ" platform for review of produced documents. We will coordinate with Sandra Neville, e-discovery project manager at Canfield, to establish expert user accounts, configure appropriate access permissions consistent with the Protective Order, and prepare curated document collections tailored to each expert's needs.

Key documents for the technical expert (Dr. Ogawa):

• GenomeDx's claim charts for all three asserted patents mapped against ProstaSignal™.

• Complete prosecution histories for the '271, '893, and '647 Patents, including all office actions, responses, amendments, and notices of allowance.

• Prior art references identified in Helios's invalidity contentions: Nakamura et al. (U.S. Patent No. 9,617,340), Gagnon (WO 2018/065413), Thornberry et al. (Journal of Computational Oncology, Vol. 14, No. 3, pp. 287–301, March 2015), and Klein et al. (European Urology, Vol. 66, No. 1, pp. 45–52, 2014).

• Helios's technical development documents relating to ProstaSignal™, including algorithm design documents, gene selection rationale, classifier training data and methodology, and validation study protocols (produced subject to the Protective Order and designated as Highly Confidential — Attorneys' Eyes Only).

• ProstaSignal™ algorithm documentation, including source code (if produced) or algorithm descriptions.

• GenomeDx's Decipher® technical documentation for comparison purposes.

• Deposition transcripts of Helios technical witnesses (as they become available).

Key documents for the damages expert (Dr. Tremaine):

• GenomeDx's audited financial statements (prepared by its independent auditors, Whitfield & Garner LLP), including income statements, balance sheets, and segment-level revenue data.

• Helios's revenue data and ProstaSignal™ test volume records (produced subject to the Protective Order).

• The Veridian Molecular Labs license agreement, dated April 3, 2022, including all amendments and exhibits.

• Market share analyses, including GenomeDx's internal market share tracking reports and any third-party market analyses identified in discovery.

• GenomeDx's Decipher® reimbursement data, including payer mix, average reimbursement per test, and trends over the relevant damages period.

• Helios investor materials, including pitch decks, financial projections, and board presentations — subject to the pending motion to compel regarding Requests for Production Nos. 22, 23, and 31, which seek documents relating to Helios's financial projections, revenue models, and investor communications. The outcome of that motion may significantly affect the data available to Dr. Tremaine.

• GenomeDx's cost data, including cost-of-goods-sold breakdowns, variable cost analysis, and incremental cost data necessary to support the incremental margin calculation.

D. Budget Considerations

Expert fees represent a significant component of the overall litigation budget for this matter. The approved litigation budget, as set forth in the engagement letter between Calderwood & Harkness LLP and GenomeDx dated March 12, 2025, is $6.8 million to $9.2 million through trial, inclusive of attorney fees, expert fees, e-discovery costs, and other litigation expenses.

Expert fees are accounted for separately from the e-discovery costs of $1.4 million to $1.8 million estimated by Canfield Document Solutions LLC for document processing, hosting, review support, and related services.

Based on preliminary discussions with both recommended experts and on the anticipated scope of their engagements, we estimate that combined expert costs (for both Dr. Ogawa and Dr. Tremaine) through trial will fall in the range of approximately $850,000 to $1,200,000. This estimate encompasses initial document review and consultation, preparation of affirmative and rebuttal expert reports, deposition preparation and testimony, Markman hearing preparation and attendance (Dr. Ogawa), trial preparation, and trial testimony. This estimate will be refined once hourly rates are confirmed in the executed engagement letters and once the scope of the experts' work is more precisely defined.

GenomeDx's intellectual property enforcement insurance policy through Greystone Specialty Insurance Co. — policy number GSI-IP-2024-07812, with $10 million in coverage, a $2.5 million deductible, and 80/20 co-insurance above the deductible — may cover a portion of the expert costs. We should coordinate with Alan Ngo, who has been handling the insurance coverage correspondence, to confirm the scope of coverage for expert fees and to ensure that all necessary claim notifications are made in a timely manner. Given the $2.5 million deductible, coverage for expert costs will depend on the total litigation expenditure exceeding that threshold, which is expected based on current projections.

V. Recommendation and Next Steps

Based on the evaluation set forth above, I recommend retaining Dr. Franklin Ogawa as GenomeDx's technical expert and Dr. Louise Tremaine as GenomeDx's damages expert. Both experts possess the qualifications, experience, and availability required to provide high-quality expert opinions and testimony in this matter, and both present strong Daubert profiles with minimal conflict risk.

Subject to your approval, the immediate next steps are as follows:

1. Finalize and circulate draft engagement letters to both Dr. Ogawa and Dr. Tremaine for their review and comment. The engagement letters will address scope of engagement, hourly rates, billing procedures, confidentiality obligations, and Protective Order compliance.

2. Conduct final conflict confirmations with both experts' organizations. Although our preliminary conflict checks have not identified any disqualifying conflicts, we should obtain formal written confirmations from Dr. Ogawa's office at Stanford and from Ashworth Economics Group LLC prior to execution of the engagement letters.

3. Obtain client approval from Victoria Sung, General Counsel of GenomeDx, for the retention of both experts, including approval of the estimated fee range. If required by the board governance provisions applicable to litigation expenditures of this magnitude, we should also inform Jonathan Hargrove at Pinnacle Biosciences Capital, LP, of the expert selections and associated costs.

4. Execute engagement letters and onboard both experts under the Protective Order. Each expert will be required to execute the Protective Order undertaking and will receive access credentials for the RevealIQ document review platform.

5. Provide initial document packages to both experts, including the curated collections of key documents described in Section IV.C above. Priority should be given to documents that are currently available, with supplemental productions to follow as additional discovery materials are produced or become available.

6. Schedule introductory calls with both experts. The introductory call with Dr. Ogawa should cover GenomeDx's infringement theories, the key claim construction disputes, the validity challenges raised by Helios, and the prosecution history issues that will inform Dr. Ogawa's opinions. The introductory call with Dr. Tremaine should cover the damages framework (lost profits and reasonable royalty), the available financial data, the Veridian comparable license, and Dr. Tremaine's preliminary data needs for constructing her damages models.

7. Coordinate with Wendy Schultz on the privilege review protocol to ensure that all expert communications — including draft reports, preliminary analyses, and attorney-expert communications — are properly protected under the work-product doctrine and, where applicable, the attorney-client privilege. This coordination is particularly important in light of the Federal Circuit's holding in In re Seagate Technology, LLC, and the evolving standards for protection of expert materials under Federal Rule of Civil Procedure 26(b)(4).

I respectfully request your approval to proceed with the retention of both recommended experts on the timeline described above. Please let me know if you have any questions or would like to discuss any aspect of this memorandum.

Peter Vassiliev Lead Associate, intellectual Property

Calderwood & Harkness LLP 555 Mission Street, Suite 3100 San Francisco, CA 94105

PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT This memorandum is protected by the attorney-client privilege and the work-product doctrine. It is intended solely for the use of the addressee and should not be disclosed to any third party without the prior written consent of Calderwood & Harkness LLP.

Milestone	Date
Expert Report Deadline	May 15, 2026
Rebuttal Expert Report Deadline	July 10, 2026
Markman Hearing	September 18, 2026
Trial	February 2027
Milestone	Date	Expert Deliverable / Activity
Expert Retention Target	February 2026	Execute engagement letters; onboard under Protective Order
Fact discovery Close	March 6, 2026	Experts begin reviewing key documents and deposition transcripts as they become available
Expert Reports Due	May 15, 2026	Affirmative expert reports (technical and damages)
Rebuttal Expert Reports Due	July 10, 2026	Rebuttal reports responding to Helios's experts
Markman Hearing	September 18, 2026	Dr. Ogawa may be needed for claim construction tutorial and/or testimony
Trial	February 2027	Both experts anticipated to testify live

MEMORANDUM
PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT
TO: Catherine Bloom, Responsible Partner, intellectual Property
FROM: Peter Vassiliev, Lead Associate, intellectual Property
DATE: January 2026
RE: Expert Selection and Qualification — Genome Dx Therapeutics Inc. v. Helios Precision Diagnostics, Inc., Case No. 3:25-cv-01847-JWL (N.D. Cal.)
CLIENT: Genome Dx Therapeutics Inc. ("GenomeDx" or "GDT")
MATTER: Patent infringement action asserting U.S. Patent Nos. 10,438,271 ("the '271 Patent"), 11,204,893 ("the '893 Patent"), and 11,891,647 ("the '647 Patent") against Helios Precision Diagnostics, Inc. ("Helios") and its ProstaSignal™ Genomic Classifier
I. Purpose and Executive Summary
This memorandum evaluates candidate technical and damages experts for plaintiff GenomeDx's affirmative patent infringement case against Helios Precision Diagnostics, Inc. and provides recommendations for the prompt retention of experts in both categories. The matter is progressing on an accelerated schedule, and early identification, vetting, and engagement of qualified experts is critical to meeting the deadlines established in the Court's Scheduling Order and to ensuring the highest quality of expert work product at the Markman hearing and at trial.
Two categories of experts are required: (1) a technical expert with deep knowledge of genomic classification methods who can opine on infringement analysis (including literal infringement and the doctrine of equivalents), patent validity in the face of Helios's invalidity contentions, and claim construction of the disputed claim terms; and (2) a damages expert with substantial experience in patent damages quantification who can present opinions on lost profits, reasonable royalty calculations, and a comprehensive Georgia-Pacific factor analysis tailored to the molecular diagnostics industry.
After evaluating multiple candidates in each category, I recommend the following:
• Technical Expert: Dr. Franklin Ogawa, PhD, Professor of Computational Oncology, Stanford University. Dr. Ogawa possesses precisely targeted expertise in machine-learning classifiers applied to genomic data, has an extensive publication record, and has testified as a technical expert in three prior patent infringement matters involving molecular diagnostics and bioinformatics — with his testimony admitted without challenge under Daubert in both the Northern District of California and the District of Delaware.
• Damages Expert: Dr. Louise Tremaine, PhD, Managing Director, Ashworth Economics Group LLC, Chicago, Illinois. Dr. Tremaine has served as a damages expert in more than twenty federal patent cases, has never had testimony excluded under Daubert, and has particular depth in life sciences, diagnostics, and medical device patent damages — including familiarity with Medicare reimbursement structures and the MolDX program, which are directly relevant to this matter.
The key case deadlines relevant to expert retention and preparation are as follows:
Draft engagement letters are being prepared concurrently for both recommended experts, and I anticipate that, subject to your approval, we can execute those letters and onboard both experts during February 2026 — providing adequate lead time for document review, consultation, and report preparation.
As background on the current posture of the case: the Court denied Helios's Motion to Dismiss under 35 U.S.C. § 101 on January 13, 2026, rejecting Helios's argument that the asserted claims are directed to patent-ineligible subject matter. The parties are currently in active fact discovery, which is scheduled to close on March 6, 2026. Claim construction briefing preparation is underway, with the Markman hearing set for September 18, 2026. The case remains on track for a February 2027 trial before Judge J. William Liang.
II. Technical Expert — Requirements and Evaluation Criteria
A. Scope of Technical Expert's Anticipated Opinions
The technical expert retained by GenomeDx will be expected to provide opinions across several interrelated areas. The scope of the anticipated opinions is summarized below and will be refined as fact discovery is completed and as the claim construction positions of both parties crystallize.
Infringement Analysis. The core of the technical expert's engagement will be an infringement analysis addressing whether Helios's ProstaSignal™ test infringes the asserted claims of the '271, '893, and '647 Patents. The ProstaSignal™ test, as currently understood from Helios's public disclosures and preliminary discovery responses, utilizes a 24-gene expression panel, processes gene expression data through a gradient-boosted decision tree classifier, and generates a ProstaScore ranging from 0 to 100 that places patients into Low, Intermediate, or High risk classifications. The expert will need to map each element of the asserted claims onto the corresponding features of the ProstaSignal™ system, with reference to both the patent specifications and the prosecution histories of each asserted patent.
Literal Infringement and Doctrine of Equivalents. The technical expert must be prepared to opine on both literal infringement and infringement under the doctrine of equivalents. The doctrine of equivalents analysis is particularly important with respect to the '893 Patent, which recites a "random-forest classifier" limitation in several of its asserted claims. Helios's ProstaSignal™ test employs a gradient-boosted decision tree classifier rather than a random-forest classifier. The expert will need to analyze whether Helios's gradient-boosted decision tree classifier performs substantially the same function, in substantially the same way, to achieve substantially the same result as the claimed random-forest classifier — or, alternatively, whether the two classifiers are insubstantially different under the Harborline Enterprises Davis formulation. This analysis will require deep technical understanding of both ensemble learning methods and the specific algorithmic and functional differences (and similarities) between random forests and gradient-boosted decision trees as applied to gene expression data.
Gene Overlap Analysis. The technical expert should also address the significance of the 19-of-24 gene overlap between ProstaSignal™'s 24-gene panel and GenomeDx's patented gene set. This overlap is relevant to multiple claim limitations across the three asserted patents, including the '271 Patent's claim limitation requiring "a panel of at least 22 genes selected from a predefined gene set." The expert will need to consider the degree of overlap, the functional significance of the shared genes, and the implications of the remaining genes in the ProstaSignal™ panel that fall outside the patented gene set.
Validity Opinions. The technical expert must be prepared to respond to Helios's invalidity contentions under 35 U.S.C. §§ 101, 102, and 103. Helios's preliminary invalidity contentions identify the following principal prior art references:
• Nakamura et al., U.S. Patent No. 9,617,340 — directed to gene expression-based classification methods for cancer prognosis, but employing a different classifier architecture and training methodology.
• Gagnon, WO 2018/065413 — an international patent publication describing multi-gene panels for urological cancer risk assessment.
• Thornberry et al., "Machine Learning Approaches to Genomic Risk Stratification in Prostate Cancer," Journal of Computational Oncology, Vol. 14, No. 3, pp. 287–301 (March 2015) — an academic publication describing early-stage research on machine-learning classifiers applied to prostate cancer gene expression data.
• Klein et al., "A 17-Gene Assay to Predict Prostate Cancer Aggressiveness in the Context of Gleason Grade Heterogeneity, Tumor Multifocality, and Biopsy Undersampling," European Urology, Vol. 66, No. 1, pp. 45–52 (2014) — describing a commercially available multi-gene assay for prostate cancer risk assessment.
The expert must be able to distinguish each of these references from the asserted claims and, where necessary, demonstrate that the claimed inventions would not have been obvious to a person of ordinary skill in the art at the time of the inventions in view of these references, whether taken individually or in combination.
Claim Construction Testimony. The technical expert may be called upon to provide testimony at the Markman hearing scheduled for September 18, 2026. The parties' joint claim construction statement identifies the following disputed claim terms, among others:
• "trained machine-learning classifier"
• "random-forest classifier"
• "panel of at least 22 genes selected from a predefined gene set"
• "risk category"
• "composite risk metric"
The expert should be able to explain to the Court how a person of ordinary skill in the art at the relevant time would understand each of these terms in the context of the patent specifications, the prosecution histories, and the state of the art in computational oncology and genomic diagnostics.
Section 101 / Patent Eligibility. Although the Court has already denied Helios's Motion to Dismiss under § 101, it is possible that Helios will reassert patent-eligibility arguments at the claim construction stage or at summary judgment. The expert must be prepared to address Helios's contention that the asserted claims are directed to abstract mathematical algorithms applied to natural phenomena (i.e., gene expression levels), citing Mayo Collaborative Services v. Prometheus Laboratories, Inc., 566 U.S. 66 (2012), and Association for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013). The expert must be able to articulate, from a technical perspective, the specific, concrete diagnostic improvements recited in the claims — including the particular combination of sample processing steps, gene selection, classifier training methodology, and clinically actionable risk categorization — that distinguish the claimed inventions from mere abstract ideas or laws of nature.
B. Desired Qualifications
In selecting a technical expert, we evaluated candidates against the following criteria, which reflect both the substantive requirements of the case and the practical demands of persuasive expert testimony:
1. Academic Credentials. PhD-level training in computational biology, bioinformatics, computational oncology, or a closely related field. An advanced degree from a recognized research institution is a baseline requirement, and active affiliation with an academic institution or major research center is strongly preferred.
2. Subject-Matter Expertise. Deep expertise in machine-learning classifiers applied to genomic data, with particular knowledge of random forests, gradient-boosted decision trees, and other ensemble methods. The candidate must be conversant with the algorithmic principles underlying these classifiers, their application to high-dimensional gene expression data, and the practical and theoretical differences between the various approaches.
3. Familiarity with Laboratory Methodologies. Working knowledge of gene expression profiling methodologies relevant to the asserted patents, including processing of formalin-fixed, paraffin-embedded (FFPE) tissue specimens, RNA extraction protocols, reference-gene normalization techniques, and the technical challenges associated with deriving reliable gene expression measurements from archived clinical specimens.
4. Clinical Context. Understanding of prostate cancer molecular diagnostics and clinical genomic risk stratification. The expert should be able to situate the technical discussion within the clinical context, explaining why and how genomic classifiers are used to inform treatment decisions for prostate cancer patients.
5. Publication Record. An active and substantial record of peer-reviewed publications in relevant journals, demonstrating recognition by the scientific community and ongoing engagement with the field.
6. Litigation Experience. Prior experience testifying as an expert witness in patent cases, preferably in the life sciences or molecular diagnostics domain. The candidate must be able to communicate complex technical concepts clearly to a judge and jury, and must have a track record that demonstrates the ability to withstand cross-examination and Daubert challenges.
7. Absence of Conflicts. No conflicts of interest with Helios, its investors (including Corinth Ventures Fund III, LP), or its outside litigation counsel (Ridgecrest Durneval LLP). Any prior relationships with parties, counsel, or key witnesses must be evaluated for potential appearance issues.
C. Candidate 1 (Recommended): Dr. Franklin Ogawa, PhD
1. Background and Qualifications. Dr. Franklin Ogawa is a Professor of Computational Oncology at Stanford University, where he has held a faculty appointment for more than fifteen years. He received his PhD in computational biology from the Massachusetts Institute of Technology and completed post-doctoral training at the Broad Institute of MIT and Harvard, where his research focused on machine-learning applications to cancer genomics — specifically, the development and validation of ensemble classifier methods for gene expression-based prognosis.
Dr. Ogawa has accumulated over twenty years of academic experience working at the intersection of bioinformatics, machine learning, and cancer diagnostics. His laboratory at Stanford is one of the leading groups in the country studying the application of computational methods to molecular oncology, and his work has directly contributed to the development and refinement of the analytical frameworks that underpin the technologies at issue in this case.
Dr. Ogawa has published more than eighty peer-reviewed articles in leading journals, including Nature Methods, Bioinformatics, the Journal of Computational Oncology, Clinical Cancer Research, European Urology, and Genome Research. His publications span a broad range of topics directly relevant to this matter, including genomic classifiers for cancer risk stratification, random forest and gradient-boosted decision tree models applied to gene expression data, and molecular risk stratification in urological cancers. Several of his publications specifically address the relative performance characteristics of random forest and gradient-boosted tree classifiers in genomic applications — a central technical issue in the '893 Patent infringement analysis.
Dr. Ogawa serves on the editorial boards of several relevant journals and is a frequent invited speaker at computational oncology conferences, including the annual meetings of the American Association for Cancer Research (AACR) and the International Society for Computational Biology (ISCB). He has served as a peer reviewer for National Institutes of Health (NIH) grants in the area of genomic diagnostics, further attesting to his standing in the field and his ability to evaluate the technical merit of diagnostic innovations.
2. Prior Testimony and Litigation Experience. Dr. Ogawa has served as a technical expert in three prior patent infringement matters involving molecular diagnostics and bioinformatics methodologies. In each of those engagements, he provided opinions on infringement, validity, and/or claim construction, and prepared detailed expert reports and tutorials for the court.
His testimony has been admitted without challenge under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), in the Northern District of California (before Judge Chen) and in the District of Delaware (before Judge Andrews). Opposing counsel in those matters did not file motions to exclude his testimony, and in one matter, the court specifically noted the thoroughness and clarity of Dr. Ogawa's expert report in its claim construction opinion.
Dr. Ogawa is experienced in preparing claim construction tutorials for Markman hearings, infringement analyses presented in claim-chart format, and validity opinions addressing both anticipation and obviousness. He has prepared materials for both bench trials and jury trials and has demonstrated an ability to explain complex technical concepts in accessible, persuasive terms.
3. Relevance to This Matter. Dr. Ogawa's qualifications are exceptionally well-suited to the technical issues presented in this case. His relevance to the specific matters at issue is summarized as follows:
First, Dr. Ogawa possesses specific expertise in the precise technical area at issue: machine-learning classifiers, including random forests and gradient-boosted decision trees, applied to genomic data derived from formalin-fixed, paraffin-embedded (FFPE) prostate cancer specimens. This is not merely adjacent expertise — it is the core of his research program and publication record.
Second, Dr. Ogawa can credibly opine on whether gradient-boosted decision tree classifiers are equivalent to random-forest classifiers under the doctrine of equivalents, which is the pivotal technical question in the '893 Patent infringement analysis. His publications comparing the two classifier families in genomic applications will provide a strong foundation for his opinions, and his academic standing will lend weight to his conclusions.
Third, Dr. Ogawa is well-positioned to address the technical significance of the 19-of-24 gene overlap between ProstaSignal™'s gene panel and GenomeDx's patented gene set. He can explain the biological and computational rationale for gene selection in panel-based classifiers and can opine on the relationship between ProstaSignal™'s 24-gene panel and the claim limitation requiring "a panel of at least 22 genes selected from a predefined gene set."
Fourth, Dr. Ogawa is well-positioned to rebut Helios's § 101 arguments by explaining, from a technical perspective, why the claimed methods represent specific, concrete improvements in diagnostic methodology — rather than abstract mathematical algorithms or laws of nature. His understanding of both the computational and the clinical dimensions of genomic classification will enable him to articulate the inventive concept embodied in the claims in terms that are both technically rigorous and accessible to the Court.
Fifth, Dr. Ogawa can address the prosecution history of the asserted patents, including: (a) the amendment made during prosecution of the '271 Patent that added specific technical steps (RNA extraction from FFPE samples and reference-gene normalization) to overcome prior art rejections; and (b) the distinction drawn during prosecution of the '893 Patent between the Nakamura reference and the claimed random-forest classifier, which was based on the training methodology and minimum cohort size requirements specified in the claims.
4. Conflict Check. We have conducted a preliminary conflict check on Dr. Ogawa, with the following results:
• Dr. Ogawa has no prior consulting or employment relationship with Helios Precision Diagnostics, Inc., Corinth Ventures Fund III, LP (Helios's lead investor), Ridgecrest Durneval LLP (Helios's outside litigation counsel), or any scientists currently or formerly affiliated with Helios.
• Dr. Ogawa has no disclosed financial interest in GenomeDx, Pinnacle Biosciences Capital, LP (GenomeDx's primary institutional investor), or Ridgeline Health Ventures.
• Dr. Ogawa has no prior professional relationship with Dr. Elaine Kossack, Helios's Chief Scientific Officer and Broad Institute alumna, or with Dr. Nathan Ashford, Helios's Chief Executive Officer, that would create an appearance of bias or partiality.
• Dr. Ogawa has co-authored one publication with Dr. Rajan Mehta, GenomeDx's Chief Technology Officer and a named inventor on all three asserted patents. The co-authored publication was a 2018 article in Bioinformatics on the application of deep learning architectures to gene expression data, approximately eight years ago. We have assessed whether this prior co-authorship could be used by the defense to challenge Dr. Ogawa's independence and conclude that a single co-authored paper from 2018, on a topic tangentially related to but distinct from the specific inventions at issue, is unlikely to create a disqualifying conflict or to provide meaningful ammunition on cross-examination. However, we recommend that this relationship be disclosed preemptively in Dr. Ogawa's expert report and, if necessary, addressed directly during his testimony to neutralize any potential line of attack.
5. Daubert Considerations. Dr. Ogawa presents an exceptionally strong profile for Daubert admissibility. His qualifications are squarely within the technical subject matter at issue, and there is no gap between his area of expertise and the opinions he will be asked to render. His methodology will be based on well-accepted principles of computational biology and machine learning, including standard analytical frameworks for comparing classifier architectures, evaluating gene panel composition, and assessing the technical significance of algorithmic choices in genomic diagnostics.
His prior judicial acceptance — specifically, the admission of his testimony without challenge in two federal districts — provides additional support for admissibility in this matter. His extensive publication record in peer-reviewed journals demonstrates that his views are recognized and credited by the relevant scientific community, further bolstering the reliability prong of the Daubert analysis.
We do not anticipate a significant risk of exclusion of Dr. Ogawa's testimony, although we will, of course, prepare a thorough Daubert defense brief should Helios move to exclude.
D. Alternative Technical Expert Candidates Considered
In addition to Dr. Ogawa, we evaluated two alternative candidates for the technical expert role. Both were ultimately not selected for the reasons described below.
1. Candidate 2: Dr. Marcia Ellison, PhD. Dr. Marcia Ellison is an Associate Professor of Bioinformatics at the University of Michigan. She holds a PhD in bioinformatics and has published approximately thirty-five peer-reviewed articles on gene expression analysis, with a focus on transcriptomic profiling in solid tumors. Her academic credentials are solid and her publication record is respectable within her subfield.
However, Dr. Ellison's qualifications are narrower than Dr. Ogawa's in several respects that are directly relevant to this matter. Most significantly, Dr. Ellison lacks specific experience with ensemble classifiers — she has not published on random forests or gradient-boosted decision trees in the context of genomic data, and her machine-learning expertise is concentrated in other methodological areas (principally neural networks and support vector machines). This gap would make it more difficult for her to provide authoritative opinions on the doctrine-of-equivalents analysis for the '893 Patent, which requires a nuanced comparison of random forests and gradient-boosted decision trees.
Additionally, Dr. Ellison has never previously testified as an expert witness in patent litigation or in any other federal proceeding. While a lack of prior testimony is not disqualifying per se, it introduces a higher degree of Daubert risk and uncertainty regarding her performance at deposition and at trial. First-time expert witnesses may be less effective in communicating with judges and jurors, and opposing counsel may attempt to exploit the absence of a track record.
No conflicts of interest were identified with respect to Dr. Ellison.
Recommendation: Not selected. Dr. Ellison's less directly relevant expertise, particularly her lack of experience with ensemble classifiers, and her absence of prior testimony experience create unnecessary risk. Dr. Ogawa is the superior candidate.
2. Candidate 3: Dr. Tobias Engström, PhD. Dr. Tobias Engström is an independent consultant and former Director of Bioinformatics at a major academic medical center. He holds a PhD in computational genomics and has a strong publication record of approximately fifty papers in genomic diagnostics, including several papers on ensemble methods in gene expression analysis. His technical qualifications are substantial, and he has testified as an expert witness on two prior occasions in molecular diagnostics patent cases — both times as a defense expert.
However, a potential issue exists with respect to Dr. Engström: he previously served as a testifying expert for Ridgecrest Durneval LLP, Helios's outside litigation counsel in this matter, in an unrelated patent infringement case approximately four years ago. While this prior engagement does not constitute a formal conflict of interest — the matters are unrelated and Dr. Engström's retention by Ridgecrest Durneval was on behalf of a different client — the relationship creates a perception of divided loyalty that could be exploited on cross-examination. Helios's counsel could argue that Dr. Engström is a "hired gun" willing to testify for either side, or could use the prior relationship to suggest that Dr. Engström has personal knowledge of Ridgecrest Durneval's litigation strategies, creating an uncomfortable dynamic.
Additionally, the fact that Dr. Engström's prior two testimonial engagements were both as a defense expert could create a narrative risk — opposing counsel might argue that Dr. Engström has now "switched sides" and is willing to offer whatever opinion his retaining counsel requests.
No other conflicts were identified.
Recommendation: Not selected. The prior relationship with defendant's counsel creates unnecessary risk, even though no formal conflict exists. The potential for cross-examination on this issue, and the perception of divided loyalty, weigh against retention. Dr. Ogawa is the superior candidate.
III. Damages Expert — Requirements and Evaluation Criteria
A. Scope of Damages Expert's Anticipated Opinions
The damages expert retained by GenomeDx will be expected to provide comprehensive opinions quantifying the damages suffered by GenomeDx as a result of Helios's infringement of the asserted patents. The anticipated scope of the damages expert's opinions encompasses the following areas:
Lost-Profits Analysis. GenomeDx's primary damages theory is based on lost profits. GenomeDx is the market leader in genomic prostate cancer prognostic testing through its Decipher® product. Prior to the launch of Helios's ProstaSignal™ test in approximately October 2024, GenomeDx held approximately 74% of the relevant market. Following the launch of ProstaSignal™, GenomeDx's market share declined to approximately 68%, representing a material erosion attributable to Helios's infringing product.
The damages expert will be expected to calculate lost profits based on the following framework, which will be refined as discovery progresses:
• Helios's estimated ProstaSignal™ test volume from October 2024 through March 2025: approximately 8,400 tests.
• Estimated number of tests diverted from GenomeDx: approximately 6,100 tests (based on the two-supplier market analysis and demand transfer theory described below).
• GenomeDx's average reimbursement per test (Decipher®): $4,210.
• Lost revenue attributable to diverted tests: approximately $25,681,000 (6,100 tests × $4,210 per test).
• GenomeDx's incremental margin on additional test volume: approximately 72%.
• Estimated lost profits: approximately $18,490,320 ($25,681,000 × 72%).
These figures are preliminary and will be updated as additional financial data is produced by both parties during the remaining weeks of fact discovery.
Two-Supplier Market Theory. To support the lost-profits claim, the expert will need to establish that GenomeDx and Helios are functionally in a two-supplier market for genomic prostate cancer prognostic testing. This analysis will require examination of the market for genomic tests that provide a clinically actionable risk classification for post-surgical prostate cancer patients, the absence of non-infringing substitutes that perform the same clinical function, and the barriers to entry in this market segment. The expert should be prepared to address the Panduit factors and to demonstrate GenomeDx's entitlement to lost profits under the established framework.
Reasonable Royalty as Alternative Damages Theory. In the event the Court determines that lost profits are not available or are not fully supported by the evidence, the damages expert will also present a reasonable royalty analysis as an alternative (or floor) damages theory. The preliminary reasonable royalty calculation is as follows:
• Royalty base: $32,088,000 (Helios ProstaSignal™ revenue from October 2024 through March 2025, calculated as 8,400 tests × $3,820 average reimbursement per test).
• Proposed royalty rate: 18%.
• Reasonable royalty: $5,775,840 ($32,088,000 × 18%).
Georgia-Pacific Factor Analysis. The reasonable royalty rate will be supported by a comprehensive analysis across all fifteen Georgia-Pacific factors, as set forth in Georgia-Pacific Corp. v. United States Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970), modified and affirmed, 446 F.2d 295 (2d Cir. 1971). Particular focus will be placed on Factor 12 (the portion of the profit or selling price that may be customary in the particular business or in comparable businesses to allow for use of the invention or analogous inventions), which provides the strongest anchor for the proposed royalty rate.
The most relevant comparable license is the license agreement between GenomeDx and Veridian Molecular Labs, dated April 3, 2022. The Veridian license has the following key terms:
• Running royalty of 15% on net revenues.
• Minimum annual royalty of $500,000.
• Licensed field limited to bladder cancer diagnostics under the '271 and '893 Patents only (the '647 Patent was not licensed to Veridian).
The damages expert will analyze the comparability of the Veridian license to the hypothetical negotiation between GenomeDx and Helios and will propose an upward adjustment of the Veridian 15% rate to 18% to account for the following distinguishing factors:
(a) Litigation context. The hypothetical negotiation is deemed to occur against the backdrop of actual infringement, whereas the Veridian license was negotiated in a non-litigation context. The Federal Circuit has recognized that a patentee may command a higher royalty rate in a litigation-context hypothetical negotiation than in a voluntary licensing transaction.
(b) Direct competitive overlap. The Veridian license was limited to bladder cancer diagnostics — a different clinical application that did not place Veridian in direct competition with GenomeDx's core prostate cancer testing business. Helios's ProstaSignal™, by contrast, directly competes with GenomeDx's Decipher® test in the prostate cancer prognostic testing market, and the resulting competitive harm justifies a premium over the Veridian royalty rate.
(c) Additional patent coverage. The Veridian license covered only the '271 and '893 Patents. The hypothetical license to Helios would encompass all three asserted patents, including the '647 Patent, which provides additional claim coverage for the genomic classification technology at issue. The inclusion of an additional patent in the hypothetical license justifies a higher royalty rate.
Enhanced Damages. The damages expert should also be prepared to address the availability of enhanced damages under 35 U.S.C. § 284. While the legal standard for enhancement is determined by the Court, the damages expert's analysis of willfulness-related financial factors — including Helios's knowledge of GenomeDx's patents prior to launching ProstaSignal™, Helios's failure to design around the asserted claims, and the degree of competitive harm to GenomeDx — may inform the Court's enhancement analysis.
Ongoing Royalty / Permanent Injunction Analysis. In the event that GenomeDx seeks a permanent injunction or, in the alternative, an ongoing royalty for post-judgment infringement, the damages expert should be prepared to address the appropriate ongoing royalty rate, taking into account the adjudication of infringement and the changed dynamics of a post-judgment negotiation.
B. Desired Qualifications
The following qualifications were used to evaluate candidate damages experts:
1. Academic Credentials. An advanced degree (PhD preferred) in economics, finance, or a closely related quantitative field.
2. Patent Damages Expertise. Extensive experience in patent damages analysis, with particular depth in the life sciences and diagnostics sector. The candidate must be thoroughly familiar with the Georgia-Pacific framework, the Panduit factors for lost-profits analysis, comparable-license analysis, apportionment principles, and other methodologies commonly employed in patent damages cases.
3. Industry Knowledge. Familiarity with the economics of the molecular diagnostics industry, including Medicare reimbursement structures (particularly the MolDX program and MAAA codes), laboratory-developed test (LDT) market dynamics, and the competitive landscape for genomic testing products.
4. Publication and Speaking Record. A strong record of publications and conference presentations in the area of patent damages and intellectual property economics, demonstrating thought leadership and peer recognition.
5. Litigation Experience. Prior testimony as a damages expert in federal patent cases, with a demonstrable track record of Daubert admissibility and persuasive performance at trial.
6. Absence of Conflicts. No conflicts of interest with Helios, Corinth Ventures Fund III, LP, or Ridgecrest Durneval LLP.
C. Candidate 1 (Recommended): Dr. Louise Tremaine, PhD
1. Background and Qualifications. Dr. Louise Tremaine is a Managing Director at Ashworth Economics Group LLC, a leading economics consulting firm based in Chicago, Illinois, that specializes in intellectual property litigation support and damages analysis. Dr. Tremaine received her PhD in economics from the University of Chicago, with a specialization in industrial organization and intellectual property valuation.
Dr. Tremaine has accumulated over fifteen years of experience as a damages expert in patent litigation, during which time she has developed particular depth in life sciences, diagnostics, and medical device patent damages. Her practice encompasses the full range of patent damages methodologies, including lost-profits analysis, reasonable royalty calculations, comparable-license analysis, apportionment, and enhanced damages assessments.
Dr. Tremaine has published articles on reasonable royalty methodologies, apportionment, and lost-profits analysis in peer-reviewed economics journals and IP law journals, including the Journal of the Patent and Trademark Office Society and Economics of Innovation and New Technology. She is a regular speaker at conferences hosted by the American intellectual Property Law Association (AIPLA) and the Licensing Executives Society (LES), where she has presented on patent damages topics including reasonable royalty determination, the evolving standards for apportionment, and the economic analysis of FRAND licensing obligations.
2. Prior Testimony and Litigation Experience. Dr. Tremaine has testified as a damages expert in more than twenty federal patent cases across a wide range of technologies and industries, with a concentration in life sciences, diagnostics, and medical devices. Her testimony has been consistently admitted under Daubert, and she has never had testimony or an expert report excluded by a court.
Notably, Dr. Tremaine has served as a damages expert for both plaintiffs and defendants in patent cases, demonstrating objectivity and analytical rigor that is not tethered to any particular advocacy position. This balanced track record enhances her credibility and makes her less vulnerable to characterization as a "plaintiff's expert" or a "defendant's expert" on cross-examination.
Of particular relevance to this matter, Dr. Tremaine has specific experience with diagnostics industry economics, including Medicare reimbursement structures under the MolDX program administered by Palmetto GBA, the assignment and use of Molecular Analysis for Therapy Choice (MAAA) procedure codes, and the market dynamics of laboratory-developed tests (LDTs). This experience is directly relevant to analyzing GenomeDx's Decipher® test (MAAA code 0047M) and Helios's ProstaSignal™ test (MAAA code 0239M), including the reimbursement rates, payer mix, and revenue recognition patterns for each product.
3. Relevance to This Matter. Dr. Tremaine is well-positioned to provide credible and persuasive damages opinions in this matter for several reasons.
First, she can credibly quantify both lost-profits and reasonable-royalty damages using GenomeDx's financial data. GenomeDx's fiscal year 2024 revenue was $387.6 million, with EBITDA of $54.3 million (a 14.0% EBITDA margin) and an average reimbursement of $4,210 per Decipher® test. Dr. Tremaine's experience with diagnostics industry economics will enable her to work efficiently with this data and to construct a damages model that accurately reflects GenomeDx's cost structure, margin profile, and incremental economics.
Second, Dr. Tremaine has extensive experience analyzing comparable licenses in patent damages cases — a critical skill for this matter, where the Veridian Molecular Labs license is the primary comparable. She is well-suited to address the comparability of the Veridian license to the hypothetical negotiation at issue and to articulate the economic rationale for the upward adjustment from 15% to 18%.
Third, Dr. Tremaine can opine on market share dynamics, including GenomeDx's decline from approximately 74% to approximately 68% market share following the launch of ProstaSignal™, and can construct a rigorous two-supplier market analysis using accepted economic frameworks.
Fourth, her familiarity with competitor-entry scenarios in molecular diagnostics markets — derived from her prior engagements in similar cases — will enable her to address the demand-transfer analysis (estimating the number of tests diverted from GenomeDx to Helios) with credibility and precision.
4. Conflict Check. We have conducted a thorough conflict check on Dr. Tremaine and Ashworth Economics Group LLC, with the following results:
• Dr. Tremaine has no prior engagement by or on behalf of Helios Precision Diagnostics, Inc., Corinth Ventures Fund III, LP, Aether Life Sciences Partners, or Ridgecrest Durneval LLP.
• Dr. Tremaine has no financial interest in GenomeDx, Pinnacle Biosciences Capital, LP, or any affiliated entities.
• Ashworth Economics Group LLC has not previously been engaged by any party adverse to GenomeDx in any matter.
• No conflicts of interest have been identified.
5. Daubert Considerations. Dr. Tremaine presents the strongest possible Daubert profile among the candidates evaluated. Her extensive track record — more than twenty federal patent damages engagements, with no instances of exclusion — provides powerful evidence that her methodologies are reliable and that her qualifications are sufficient to support the opinions she will be asked to render.
The methodologies she employs — including the Georgia-Pacific framework, the Panduit lost-profits test, and comparable-license analysis — are well-established and routinely accepted by federal courts. Her academic credentials (PhD in economics from the University of Chicago) and her publication record in peer-reviewed economics and IP law journals further support the reliability of her opinions under the Daubert standard.
We do not anticipate a significant risk of exclusion of Dr. Tremaine's testimony.
D. Alternative Damages Expert Candidates Considered
In addition to Dr. Tremaine, we evaluated two alternative candidates for the damages expert role. Both were ultimately not selected for the reasons described below.
1. Candidate 2: Dr. Richard Halstead, PhD. Dr. Richard Halstead is a Principal at Centennial Valuation Partners LLC in New York, New York. He holds a PhD in finance and has strong quantitative credentials, with particular expertise in financial modeling and valuation. Dr. Halstead has served as a patent damages expert in approximately ten prior engagements.
However, the substantial majority of Dr. Halstead's patent damages experience has been in the semiconductor and software industries. He has limited experience in the life sciences and diagnostics sectors and has not previously been retained in a case involving genomic testing, Medicare reimbursement, or MolDX program economics. Retaining Dr. Halstead would require significant investment of time and resources to educate him on the genomic testing market, Medicare reimbursement structures, MAAA codes, and the specific competitive dynamics of the prostate cancer diagnostics market.
This knowledge gap creates two related risks. First, there is a heightened risk that Helios could challenge Dr. Halstead's qualifications under Daubert on the grounds that he lacks sufficient knowledge of the industry to render reliable opinions. While this challenge might not ultimately succeed, it would consume litigation resources and could create unnecessary uncertainty. Second, the steeper learning curve would increase the overall cost of expert engagement and the time required to prepare the expert report, compressing an already tight timeline.
No conflicts of interest were identified with respect to Dr. Halstead.
Recommendation: Not selected. Dr. Halstead's lack of life sciences industry experience creates risk that his opinions may not withstand Daubert challenge or may be less persuasive to the jury, and the steeper learning curve increases costs and timeline pressure.
2. Candidate 3: Dr. Karen Nussbaum, PhD. Dr. Karen Nussbaum is an independent economics consultant based in Washington, DC, with extensive experience in life sciences patent damages. She has served as a damages expert in approximately fifteen prior matters, many of which involved pharmaceutical, biotechnology, or diagnostics patents. Her qualifications and industry experience are comparable to Dr. Tremaine's, and she would be a strong candidate in the abstract.
However, Dr. Nussbaum is currently retained as a damages expert in two other active patent cases with trial dates in 2026. Given the volume of work required for expert report preparation, deposition, and trial testimony in those concurrent matters, there are serious concerns about Dr. Nussbaum's availability for the May 15, 2026 expert report deadline and the February 2027 trial in this case. Dr. Nussbaum's office has indicated that she could commit to the engagement, but has acknowledged that scheduling would be "challenging" and that some work might need to be delegated to other professionals at her firm — a less-than-reassuring response given the complexity and financial significance of this matter.
No conflicts of interest were identified with respect to Dr. Nussbaum.
Recommendation: Not selected. Scheduling constraints create an unacceptable risk of delay or inadequate preparation. The expert report deadline is firm, and GenomeDx cannot afford a damages expert who is overextended across multiple matters. Dr. Tremaine is the superior candidate.
IV. Engagement Structure and Logistics
A. Retention Terms
Both recommended experts — Dr. Franklin Ogawa (technical) and Dr. Louise Tremaine (damages) — will be retained through Calderwood & Harkness LLP on behalf of GenomeDx. Standard expert engagement letters are being drafted for each expert and will be circulated for review promptly upon approval of the expert selections. Separate draft engagement letters are being prepared for each expert and will be provided under separate cover.
Both experts will be subject to the Stipulated Protective Order entered by Judge Liang on December 18, 2025, including the "Highly Confidential — Attorneys' Eyes Only" designation tier, which will govern their access to ProstaSignal™ algorithm details, Helios's financial data, and GenomeDx's proprietary information. Each engagement letter will include an acknowledgment by the expert of the Protective Order's terms, and each expert will execute the Protective Order's required undertaking prior to receiving any confidential or highly confidential materials.
Hourly rates for each expert will be negotiated during the engagement letter process. Based on preliminary discussions, we anticipate that Dr. Ogawa's rate will be in the range of $800–$950 per hour and Dr. Tremaine's rate will be in the range of $900–$1,100 per hour, consistent with prevailing market rates for experts of their caliber in the Northern District of California.
B. Timeline and Deliverables
The retention target for both experts is February 2026, with engagement letters to be executed promptly following client approval and final conflict confirmations. The following table summarizes the key milestones and the expert-related deliverables associated with each:
Given the expert report deadline of May 15, 2026, and the volume of material that both experts will need to review and analyze, it is critical that we onboard the experts as early as possible in February 2026 and begin providing them with initial document packages and case background materials without delay.
C. Document Access and Coordination
Both experts will receive access to the Canfield Document Solutions LLC "RevealIQ" platform for review of produced documents. We will coordinate with Sandra Neville, e-discovery project manager at Canfield, to establish expert user accounts, configure appropriate access permissions consistent with the Protective Order, and prepare curated document collections tailored to each expert's needs.
Key documents for the technical expert (Dr. Ogawa):
• GenomeDx's claim charts for all three asserted patents mapped against ProstaSignal™.
• Complete prosecution histories for the '271, '893, and '647 Patents, including all office actions, responses, amendments, and notices of allowance.
• Prior art references identified in Helios's invalidity contentions: Nakamura et al. (U.S. Patent No. 9,617,340), Gagnon (WO 2018/065413), Thornberry et al. (Journal of Computational Oncology, Vol. 14, No. 3, pp. 287–301, March 2015), and Klein et al. (European Urology, Vol. 66, No. 1, pp. 45–52, 2014).
• Helios's technical development documents relating to ProstaSignal™, including algorithm design documents, gene selection rationale, classifier training data and methodology, and validation study protocols (produced subject to the Protective Order and designated as Highly Confidential — Attorneys' Eyes Only).
• ProstaSignal™ algorithm documentation, including source code (if produced) or algorithm descriptions.
• GenomeDx's Decipher® technical documentation for comparison purposes.
• Deposition transcripts of Helios technical witnesses (as they become available).
Key documents for the damages expert (Dr. Tremaine):
• GenomeDx's audited financial statements (prepared by its independent auditors, Whitfield & Garner LLP), including income statements, balance sheets, and segment-level revenue data.
• Helios's revenue data and ProstaSignal™ test volume records (produced subject to the Protective Order).
• The Veridian Molecular Labs license agreement, dated April 3, 2022, including all amendments and exhibits.
• Market share analyses, including GenomeDx's internal market share tracking reports and any third-party market analyses identified in discovery.
• GenomeDx's Decipher® reimbursement data, including payer mix, average reimbursement per test, and trends over the relevant damages period.
• Helios investor materials, including pitch decks, financial projections, and board presentations — subject to the pending motion to compel regarding Requests for Production Nos. 22, 23, and 31, which seek documents relating to Helios's financial projections, revenue models, and investor communications. The outcome of that motion may significantly affect the data available to Dr. Tremaine.
• GenomeDx's cost data, including cost-of-goods-sold breakdowns, variable cost analysis, and incremental cost data necessary to support the incremental margin calculation.
D. Budget Considerations
Expert fees represent a significant component of the overall litigation budget for this matter. The approved litigation budget, as set forth in the engagement letter between Calderwood & Harkness LLP and GenomeDx dated March 12, 2025, is $6.8 million to $9.2 million through trial, inclusive of attorney fees, expert fees, e-discovery costs, and other litigation expenses.
Expert fees are accounted for separately from the e-discovery costs of $1.4 million to $1.8 million estimated by Canfield Document Solutions LLC for document processing, hosting, review support, and related services.
Based on preliminary discussions with both recommended experts and on the anticipated scope of their engagements, we estimate that combined expert costs (for both Dr. Ogawa and Dr. Tremaine) through trial will fall in the range of approximately $850,000 to $1,200,000. This estimate encompasses initial document review and consultation, preparation of affirmative and rebuttal expert reports, deposition preparation and testimony, Markman hearing preparation and attendance (Dr. Ogawa), trial preparation, and trial testimony. This estimate will be refined once hourly rates are confirmed in the executed engagement letters and once the scope of the experts' work is more precisely defined.
GenomeDx's intellectual property enforcement insurance policy through Greystone Specialty Insurance Co. — policy number GSI-IP-2024-07812, with $10 million in coverage, a $2.5 million deductible, and 80/20 co-insurance above the deductible — may cover a portion of the expert costs. We should coordinate with Alan Ngo, who has been handling the insurance coverage correspondence, to confirm the scope of coverage for expert fees and to ensure that all necessary claim notifications are made in a timely manner. Given the $2.5 million deductible, coverage for expert costs will depend on the total litigation expenditure exceeding that threshold, which is expected based on current projections.
V. Recommendation and Next Steps
Based on the evaluation set forth above, I recommend retaining Dr. Franklin Ogawa as GenomeDx's technical expert and Dr. Louise Tremaine as GenomeDx's damages expert. Both experts possess the qualifications, experience, and availability required to provide high-quality expert opinions and testimony in this matter, and both present strong Daubert profiles with minimal conflict risk.
Subject to your approval, the immediate next steps are as follows:
1. Finalize and circulate draft engagement letters to both Dr. Ogawa and Dr. Tremaine for their review and comment. The engagement letters will address scope of engagement, hourly rates, billing procedures, confidentiality obligations, and Protective Order compliance.
2. Conduct final conflict confirmations with both experts' organizations. Although our preliminary conflict checks have not identified any disqualifying conflicts, we should obtain formal written confirmations from Dr. Ogawa's office at Stanford and from Ashworth Economics Group LLC prior to execution of the engagement letters.
3. Obtain client approval from Victoria Sung, General Counsel of GenomeDx, for the retention of both experts, including approval of the estimated fee range. If required by the board governance provisions applicable to litigation expenditures of this magnitude, we should also inform Jonathan Hargrove at Pinnacle Biosciences Capital, LP, of the expert selections and associated costs.
4. Execute engagement letters and onboard both experts under the Protective Order. Each expert will be required to execute the Protective Order undertaking and will receive access credentials for the RevealIQ document review platform.
5. Provide initial document packages to both experts, including the curated collections of key documents described in Section IV.C above. Priority should be given to documents that are currently available, with supplemental productions to follow as additional discovery materials are produced or become available.
6. Schedule introductory calls with both experts. The introductory call with Dr. Ogawa should cover GenomeDx's infringement theories, the key claim construction disputes, the validity challenges raised by Helios, and the prosecution history issues that will inform Dr. Ogawa's opinions. The introductory call with Dr. Tremaine should cover the damages framework (lost profits and reasonable royalty), the available financial data, the Veridian comparable license, and Dr. Tremaine's preliminary data needs for constructing her damages models.
7. Coordinate with Wendy Schultz on the privilege review protocol to ensure that all expert communications — including draft reports, preliminary analyses, and attorney-expert communications — are properly protected under the work-product doctrine and, where applicable, the attorney-client privilege. This coordination is particularly important in light of the Federal Circuit's holding in In re Seagate Technology, LLC, and the evolving standards for protection of expert materials under Federal Rule of Civil Procedure 26(b)(4).
I respectfully request your approval to proceed with the retention of both recommended experts on the timeline described above. Please let me know if you have any questions or would like to discuss any aspect of this memorandum.
Peter Vassiliev Lead Associate, intellectual Property
Calderwood & Harkness LLP 555 Mission Street, Suite 3100 San Francisco, CA 94105
PRIVILEGED AND CONFIDENTIAL — ATTORNEY WORK PRODUCT This memorandum is protected by the attorney-client privilege and the work-product doctrine. It is intended solely for the use of the addressee and should not be disclosed to any third party without the prior written consent of Calderwood & Harkness LLP.
Milestone | Date
Expert Report Deadline | May 15, 2026
Rebuttal Expert Report Deadline | July 10, 2026
Markman Hearing | September 18, 2026
Trial | February 2027
Milestone | Date | Expert Deliverable / Activity
Expert Retention Target | February 2026 | Execute engagement letters; onboard under Protective Order
Fact discovery Close | March 6, 2026 | Experts begin reviewing key documents and deposition transcripts as they become available
Expert Reports Due | May 15, 2026 | Affirmative expert reports (technical and damages)
Rebuttal Expert Reports Due | July 10, 2026 | Rebuttal reports responding to Helios's experts
Markman Hearing | September 18, 2026 | Dr. Ogawa may be needed for claim construction tutorial and/or testimony
Trial | February 2027 | Both experts anticipated to testify live

MEMORANDUM

Record

Parties

Dates

Amounts