Abstract
A companion diagnostic is a diagnostic test that provides information essential for the safe and effective use of a corresponding therapeutic product. To obtain marketing approval, the companion diagnostic must demonstrate acceptable analytical and clinical performance. Companion diagnostic regulations are intended to protect patients by ensuring quality and consistency of treatment-guiding biomarker testing in clinical trials and clinical practice. However, current regulations have had unintended negative consequences relating to innovation, implementation, and accessibility of precision medicine; increasing complexity and cost burden; and inhibiting development of novel diagnostics and biomarker-targeted therapeutics. We propose a range of practical solutions to these challenges, advocating that regulators, pharmaceutical companies, molecular pathologist groups, and diagnostic companies work together to increase flexibility and promote diagnostic innovation, while maintaining high-quality diagnostic testing to ensure all patients get the most appropriate treatments.
Introduction
In many regions, including the Unites States, the European Union, Japan, and South Korea, the regulatory process for approving novel therapies in a biomarker-selected population is aligned with the development of a companion diagnostic, defined as a diagnostic test that provides information essential for the safe and effective use of a corresponding therapeutic product. To obtain marketing approval, the companion diagnostic must demonstrate acceptable analytic performance and clinical validity (1–4). This approach has ensured consistency and reproducibility with respect to biomarker testing in pivotal clinical trials, and other medical agencies are aligning toward a similar model (5). However, the one-drug/one-test model has created unintended barriers to diagnostic innovation.
Challenges and Barriers Caused by Companion Diagnostic Regulations, with Examples
To comply with diagnostic regulations, pharmaceutical companies have partnered separately with diagnostic companies to co-develop different companion diagnostics for the same biomarker. Where multiple drugs (and associated diagnostics) are approved within the same cancer type, this could lead to a patient being tested more than once for the same biomarker. For example, in the field of PD-L1 testing for immunotherapies, four different tests have been approved (Table 1), with different expression cut-offs or scoring algorithms [e.g., immune cell (IC) expression, tumor cell (TC) expression, or a combination of both] depending on the drug and/or indication. Prevalence rates of biomarker-selected populations may therefore vary for the same biomarker. In non–small cell lung cancer (NSCLC), depending on the setting, a patient may require PD-L1 testing using cut-offs of TC ≥ 1%, TC ≥ 50%, or an algorithm of “TC ≥ 50 or IC ≥ 10%.”
Table 1.
Approved PD-L1 companion diagnostics in the United States (US), European Union (EU), Japan, and China.
| PD-L1 companion diagnostic | Drug | US | Japan | EUa | China | ||||
|---|---|---|---|---|---|---|---|---|---|
| Indication | Cut-off | Indication | Cut-off | Indication | Cut-off | Indication | Cut-off | ||
| PD-L1 IHC 22C3 pharmDx (Agilent) | Pembrolizumab | NSCLC | TPS ≥1% | NSCLC | TPS ≥1%, TPS ≥50% | NSCLC | TPS ≥1%, TPS ≥50% | NSCLC | TPS ≥1%, TPS ≥ 50% |
| TNBC | CPS ≥10 | TNBC | CPS≥10 | Bladder | CPS ≥1 | HNSCC | CPS ≥20 | ||
| Gastric | CPS ≥1 | TNBC | CPS ≥10 | ||||||
| GEJ | CPS ≥1, CPS ≥10 | GEJ | CPS ≥10 | ||||||
| Esophageal | CPS ≥10 | Esophageal | CPS ≥10 | ||||||
| Cervical | CPS ≥1 | Cervical | CPS ≥1 | ||||||
| HNSCC | CPS ≥1 | HNSCC | CPS ≥1, TPS ≥50 | ||||||
| Cemiplimab | NSCLC | TC ≥50% | NSCLC | TC ≥50% | NSCLC | ||||
| PD-L1 IHC 28-8 pharmDx (Agilent) | Nivolumab ± Ipilimumab | NSCLC | TC ≥ 1% | NSCLC | TC ≥1% | NSCLC | TC ≥1% | ||
| Bladder | TC ≥1 | HNSCC | TC ≥1% | ||||||
| Gastric | CPS ≥5 | ||||||||
| GEJ | CPS ≥5 | ||||||||
| Esophageal | CPS ≥1, CPS ≥5 | ||||||||
| VENTANA PD-L1 (SP263) assay (VENTANA) | Durvalumab | NSCLC | TC ≥1% | ||||||
| Atezolizumab | NSCLC | TC ≥1% | NSCLC | TC ≥1% | NSCLC | TC ≥50% | NSCLC | ||
| Cemiplimab | NSCLC | TC ≥50% | NSCLC | TC ≥1%, TC ≥50% | |||||
| Tislelizumab | Bladder | TC ≥25% or IC ≥25% | |||||||
| VENTANA PD-L1 (SP142) assay (VENTANA) | Atezolizumab | NSCLC | TC ≥50 or IC ≥10% | NSCLC | TC ≥50 or IC ≥10% | NSCLC | TC ≥50% or IC ≥10% | NSCLC | TC ≥50% or IC ≥10% |
| Bladder | IC ≥5% | Bladder | IC ≥5% | ||||||
| TNBC | IC ≥1% | ||||||||
Note: The US and Japan have also approved PD-L1 complementary diagnostics, which are not included here.
China cut-off data obtained from diagnostic instructions for use. China has also approved AmoyDx PD-L1 (E1L3N) assay for pembrolizumab in NSCLC, MEDx E1L3N Antibody Kit for nivolumab in NSCLC and WuXi Diagnostics WD160 assay for zimberelimab in cervical cancer.
Data correct as of August 07, 2024.
Abbreviations: CPS, combined positive score; GEJ, gastroesophageal junction; HNSCC, head and neck squamous cell carcinoma; NSCLC, non–small cell lung cancer; TNBC, triple negative breast cancer; TPS, tumor proportion score.
The EU is currently transitioning to in vitro diagnostic regulations (IVDR), where diagnostics need to be formally approved; this table anticipates approval that will be required within IVDR based on current EU approvals for PD-L1–selected populations.
Requiring laboratories to offer multiple tests with multiple unique cut-offs for the same biomarker is resource intensive, challenging to pathologists, who may not know which drug is being offered to patients, and confusing to clinicians, who may not know which test to order (6). It also increases the burden on the payer, with reimbursement potentially required for more than one test for the same biomarker. Not all laboratories have the resources or instrumentation to provide every companion diagnostic test; in a 2022 survey performed on behalf of AstraZeneca, only three of the top 20 US clinical laboratories by market share for NSCLC testing offered all four commercially available PD-L1 companion diagnostic tests (Diaceutics, personal communication). For practical reasons, most laboratories will validate and perform a single test for any biomarker.
If different diagnostic assays identify similar but nonidentical patient populations, multiple “biomarker-positive” indications may be created within a tumor type, or even across multiple tumors, increasing complexity and making it challenging to assess the relative performance of therapeutics. Homologous recombination deficiency (HRD) is a phenotype whereby a cell cannot effectively repair DNA double-strand breaks. The presence of HRD is associated with improved response to PARP inhibitors and platinum chemotherapy. Germline mutations in BRCA1 and BRCA2 are well-established as leading to HRD. However, additional genetic and epigenetic changes (e.g., alterations to ATM, PALB2, and RAD51) are also associated with an HRD phenotype. Signatures of genomic instability associated with HRD include LOH and telomeric imbalances, and chromosome translocations, inversions, or deletions. There are several approved companion diagnostics that facilitate patient selection for PARP inhibitor therapy. FoundationOne CDx and Myriad MyChoice CDx assess chromosomal instability, whereas other FDA-approved assays only detect genetic aberrations associated with HRD (Myriad BRACAnalysis CDx and FoundationOne Liquid CDx). Additionally, variations in assay performance, bioinformatics pipelines, cut-offs, and reporting styles mean that each assay takes a different approach to reporting a sample as HRD-positive versus HRD-negative (7).
Confusion has also arisen from diagnostic labeling. Erdafitinib was FDA-approved in 2019 for metastatic bladder cancer in patients with “susceptible FGFR3 or FGFR2 genetic alterations.” The approved companion diagnostic test detects four specific FGFR2/3 mutations and five specific translocations, but only two fusions form part of the indications for use for the FDA diagnostic approval. In contrast, four FGFR2/3 fusions are mentioned in the erdafitinib US Prescribing Information (refs. 8, 9). As a result, interpretation of what constitutes susceptible genetic alterations may vary between testing centers, especially as information on new variants comes to light. Current regulations do not allow enough flexibility for the incorporation of emerging data, especially in cases where an approved assay holds a market-dominant position.
Once a drug is approved in a biomarker-selected population, it is unethical to use a placebo, which limits the ability to statistically assess the predictive performance of a subsequent companion diagnostic without access to the tissue from the original trial. Even if there is compelling retrospective data showing new assays or algorithms can better identify responding patients, a prospective clinical trial would be required for approval of a new companion diagnostic. Addressing these challenges may require newer statistical approaches. Where standard-of-care (SoC) therapy is established in a biomarker subgroup, subsequent prospective clinical trials of innovative medicines must be performed in the same population. Consequently, all subsequent trials would be conducted using the approved assay to allow assessment of the novel treatment versus SoC and ensure patients randomized to the SoC arm are treated per the approved criteria (Fig. 1A). Indeed, the FDA recommends this to pharmaceutical companies during investigational new drug application discussions. Yet, under current companion diagnostic regulations, mandating a specific assay is problematic. Any pharmaceutical company seeking FDA approval of a novel therapeutic in a biomarker-selected population is required to submit a pre-market authorization in partnership with a diagnostic manufacturer. If the diagnostic company refuses partnership, the novel therapeutic cannot gain with the approved companion diagnostic, and the pharmaceutical company will need to find another diagnostic strategy. A diagnostic company may refuse partnership for a number of reasons. They may have entered into an exclusive commercial relationship with a pharma partner that precludes working with third parties. In other cases, production capacity may only be sufficient to meet the existing market demand, and the business case to expand without a guarantee of future sales is not as compelling as other investment opportunities. In our experience, diagnostic partners who work freely with third parties are more likely to see their tests spread widely in the market and to encourage innovation.
Figure 1.
Clinical trial design options in biomarker-positive populations defined by a companion diagnostic test. A, Clinical design of innovative therapies encouraged by current regulations. B, Removal of supplemental pre-market authorization requirement. C, Flexible design allows diagnostic innovation. SoC, standard of care; PMA, premarket approval; sPMA, supplemental premarket approval.
In many instances, companion diagnostic tests are expensive, making it difficult for commercial laboratories to offer the test. Diagnostic cost/reimbursement is an important challenge and often serves as a roadblock to clinical adoption, even where clinical data clearly support diagnostic testing (10). In practice, testing is often done via laboratory developed tests (LDT). The use of LDTs can broaden access to testing and reduce costs. However, the FDA has recently increased LDT oversight (11), due to concerns that LDTs may not be as robust as the approved companion diagnostic, despite evidence of excellent performance for single genetic mutation and next-generation sequencing LDTs (12, 13). In the European Union, LDTs with a medical purpose must comply with the in vitro diagnostic regulations.
How Can These Challenges Be Addressed?
Assay concordance studies
The challenges of the current companion diagnostic landscape have been recognized by the FDA in a recent publication (14). Where multiple assays are approved, concordance data between different assays may assist testing labs in understanding if they need to validate multiple assays or can use a single test. Cross-party PD-L1 harmonization initiatives demonstrated good concordance between three of the four approved PD-L1 assays, at least when comparing TC staining in NSCLC (15, 16). However, concordance is poorer in other cancer indications, such as urothelial and breast cancers (17–19). Higher inter-assay discordance is frequently associated with the inclusion of IC PD-L1 expression within the scoring algorithms, and even for TC staining, different studies have found a range of concordance between PD-L1 assays (20). These apparent contradictions may be explained by differences in PD-L1 epitopes recognized by the antibody clones. Assays such as PD-L1 IHC 22C3 pharmDx and PD-L1 IHC 28-8 pharmDx recognize external epitopes and are more susceptible to loss during nonoptimal fixation and biopsy handling. If samples were not optimally fixed or fixation was delayed, these assays showed poor concordance with the more robust internal epitope assays. Under optimal fixation conditions, concordance was high (21). These data emphasize the importance of correct sample handling and processing in diagnostic testing.
Other IHC assays are likely to face similar challenges. The recent approval of trastuzumab deruxtecan in HER2-low breast cancer with an associated companion diagnostic [VENTANA anti-HER2/neu (4B5) Rabbit Monoclonal Primary Antibody] will raise questions about the performance of other HER2 assays at this new threshold and how this patient population should be defined in future clinical studies. Robust concordance data between the regulatory approved companion diagnostic assay and other on-market assays at the HER2-low cut-off will be vital in understanding whether assays can be used interchangeably. If samples and resources were not constrained, pivotal trials could include efficacy data of patients tested with all approved diagnostic assays, as this will provide the best evidence as to whether the assays are comparable in describing patient benefit. However, there is a limit to tissue and blood samples available from clinical studies (patient safety being paramount). For simpler diagnostics (e.g., single target genetic mutations), analytical performance data for new assays for existing biomarkers could be considered as a pathway to broaden access to testing in the absence of clinical data or where the approved gold standard assay cannot be accessed for concordance studies.
Changes to diagnostic labeling
Consultation with pathology associations during drug labeling discussions may improve clarity on what constitutes the biomarker-selected population. To enable broad access to testing in the market while also ensuring quality, we recommend regulators align on wording in drug labels that requires the use of an “analytically validated test,” rather than an “approved test.” The FDA has proposed a companion diagnostic pilot study (22) with an aim of establishing minimal performance criteria, which could be a step toward a simplified solution that enables broader access to testing once the drug is approved. Clear guidance on what constitutes a validated test must be provided, which could be aligned with published recommendations from respected expert groups (23, 24).
Another option is for regulators to approve companion diagnostics for a drug class, rather than individual drugs. The FDA has taken this approach already in a limited number of cases (25); Japan also has drug agnostic companion diagnostic guidance (26). For simpler diagnostics such as those detecting EGFR mutations, this solution has proved practical and effective. However, it is unclear whether this approach is appropriate for more complex assays, particularly where different assays do not perform in exactly the same way, such as the HRD and PD-L1 assays discussed previously. In January 2024, the FDA announced a reclassification process whereby companion diagnostics may be approved via the less burdensome 510(k) process. It is currently unclear how this will work in practice and whether it will improve testing access (27).
Generation of efficacy data for patients selected by different diagnostic assays
Pharmaceutical companies could be encouraged to generate efficacy information in pivotal clinical studies not only with their own partnered diagnostic assay but also with any previously approved diagnostic assay(s) in that setting. Although there will be additional costs associated with such an undertaking, these data would be of clear value to the clinical community. Quality of testing could be assured by using the approved assay in laboratories with ISO 15189 accreditation [or College of American Pathologists (CAP)/Clinical Laboratory Improvement Amendments (CLIA) in the United States] according to its instructions for use and generating efficacy data under a pre-specified statistical analysis plan (which could be first discussed and agreed with regulators). Testing could include a significant proportion of the randomized study population. These data could be provided after approval and made accessible to the clinical community, perhaps by being held by regulators on a publicly available website.
Flexibility in conduct of clinical trials in selected populations defined by an approved companion diagnostic test
To gain regulatory approval, newer therapeutics require direct comparison with existing agents, including in a selected population (14). The potential options for registrational intent studies in biomarker-selected populations are outlined in Fig. 1. Panel A may represent the simplest solution, but, as indicated above, this is problematic if drug developers cannot secure access to the approved diagnostic. Diagnostic companies could recommend that any exclusivity provisions are time-limited, to avoid the risk of stifling innovation. Thus, after a certain time period has expired, access to the assay for regulatory submission purposes will be available to all. Another solution to this problem is to remove the requirement for a supplemental pre-market authorization/diagnostic approval in a scenario in which an approved on-market test, performed according to College of American Pathologists/CLIA (or relevant local quality criteria), is used for patient selection in a pivotal trial (Fig. 1B). The assay could therefore be purchased and used “off the shelf,” and provided the assay is used in accredited laboratories according to the manufacturer’s instructions, no additional regulatory submissions would be needed. This would remove the potential for diagnostic manufacturers and/or pharma companies to block or otherwise impede drug approvals by denial of partnership.
Specifying the use of approved assays also prevents diagnostic innovation. As stated above, approved HRD tests now define SoC populations in certain types of ovarian, prostate, and pancreatic cancers. However, there is ongoing debate on the best way to define HRD, with newer assays showing potentially better performance (7). Ideally, algorithms would evolve over time in response to new evidence. However, if a new, innovative assay is used for patient selection in clinical trials, there is a risk that patients in the SoC arm are not treated according to the approved labeling, particularly if the novel assay is nonidentical to the approved assay (Fig. 1C). For this approach to be viable to regulators, it is likely that acceptable risk–benefit would need to be established in phase I/II clinical studies and would be most relevant where there is evidence that the approved diagnostic is not optimal in identifying patients likely to respond to a specific drug class. We would also recommend patients in the pivotal clinical trial and/or the preceding phase II trial be tested in parallel with the approved assay in addition to the new assay to enable efficacy data to be generated. These data would provide the best evidence of clinical utility for each assay. Assay concordance data should also be generated and published to inform the scientific community.
Another flexible option would be to permit the use of clinical trial assays and/or “in-house” assays for patient selection in pivotal trials with a potential path to drug approval without a companion diagnostic. The FDA companion diagnostic pilot study could be a step in this direction. It would mean that any test meeting the established minimum performance criteria can be used for patient selection without requiring formal approval. However, the FDA pilot study is currently only open to limited, high unmet need populations. Also, there is a requirement to generate performance data against a gold standard reference assay. This would require right of access to analytic data for the gold standard assay which may not be granted by the assay manufacturer. The authors note that a handful of drugs have been FDA-approved in biomarker-selected populations without a companion diagnostic (28). In such cases, the FDA has required a post-marketing commitment to develop a companion diagnostic.
Summary and Conclusions
Challenges and recommendations are summarized in Table 2. There are opportunities for regulators, pharmaceutical companies, molecular pathologist groups, and diagnostic companies to work together to increase flexibility and promote diagnostic innovation, while ensuring high-quality diagnostic testing in clinical trials and in clinical practice. Language that permits test flexibility, provided the assay is appropriately validated, is a key step to providing an environment more conducive to increased access to personalized medicine.
Table 2.
Summary of current challenges to companion diagnostic implementation and recommendations for change.
| Challenge | Recommendation |
|---|---|
| Confusion caused by approval of multiple assays for single biomarker |
|
| Confusion caused by “approved assay” terminology in drug labeling |
|
| Lack of clarity on biomarker selection criteria in drug label | Consultation with recognized academic associations involved in biomarker testing, e.g., Association for Molecular Pathology (AMP), International Association for the Study of Lung Cancer (IASLC), College of American Pathologists (CAP), and European Society of Pathology (ESP) prior to labeling discussions |
| Access to approved assay for drug/diagnostic co-registration not granted | Permit approved companion diagnostics to be used “off the shelf” in approved indication with no requirement for partnership between a diagnostic supplier and a pharmaceutical company. Remove requirement for supplemental PMA. All testing should be in compliance with CAP/CLIA (or equivalent outside the United States) |
| Lack of innovation in diagnostics due to “first past the post” diagnostic defining intended use population | Permit the use of new diagnostics in phase III studies, even for SoC arm. Ideally, the risk/benefit would be informed by clinical data, with efficacy data for the approved diagnostic included in the pivotal study |
Acknowledgments
We thank Henry Zhou and Kosho Murayama for providing information on companion diagnostic approvals in China and Japan, respectively.
Authors’ Disclosures
K.S. Oliner reports other support from Elephas Biosciences Corporation outside the submitted work. M. Shiller reports other support from AstraZeneca, Eli Lilly and Company, Astellas, Pfizer, Merck, Amgen, Mirati, AbbVie, Roche, and Bristol Myers Squibb outside the submitted work. P. Schmid reports grants from Astellas, Genentech, Oncogenex, and Medivation; grants and personal fees from AstraZeneca, Merck, Novartis, and Roche; and personal fees from Bayer, Boehringer Ingelheim, Pfizer, Puma, Eisai, and Celgene outside the submitted work. M.J. Ratcliffe reports personal fees from AstraZeneca outside the submitted work. A.J. Schetter reports other support from AstraZeneca outside the submitted work. M.-S. Tsao reports grants and personal fees from AstraZeneca, Sanofi, and Bayer and personal fees from AbbVie, Amgen, and Daichii Sankyo outside the submitted work.
References
- 1. FDA . Principles for codevelopment of an in vitro companion diagnostic device with a therapeutic product. Bethedsa (MD): U.S. Food and Drug Administration; 2018[cited 2024 Aug]. Available from:https://www.fda.gov/regulatory-information/search-fda-guidance-documents/principles-codevelopment-in-vitro-companion-diagnostic-device-therapeutic-product. [Google Scholar]
- 2. MFDS . Act on in vitro diagnostic medical devices, law 163433 online2020 [cited 2024 Jun]. Available from:https://www.mfds.go.kr/eng/index.do.
- 3. PMDA . Technical guidance on development of in vitro companion diagnostics and corresponding therapeutic products; 2013[cited 2024 Jun]. Available from:https://www.pmda.go.jp/files/000153149.pdf.
- 4. EU . Regulation (EU) 2017/746 of the European Parliament and of the Council of 5 April 2017 on in vitro diagnostic medical devices and repealing Directive 98/79/EC and Commission Decision 2010/227/EU 2023[cited 2024 Aug]. Available from:https://eur-lex.europa.eu/eli/reg/2017/746/2023-03-20.
- 5. Valla V, Alzabin S, Koukoura A, Lewis A, Nielsen AA, Vassiliadis E. Companion diagnostics: state of the art and new regulations. Biomark Insights 2021;16:11772719211047763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Salgado R, Bellizzi AM, Rimm D, Bartlett JMS, Nielsen T, Holger M, et al. How current assay approval policies are leading to unintended imprecision medicine. Lancet Oncol 2020;21:1399–401. [DOI] [PubMed] [Google Scholar]
- 7. Perez-Villatoro F, Oikkonen J, Casado J, Chernenko A, Gulhan DC, Tumiati M, et al. Optimized detection of homologous recombination deficiency improves the prediction of clinical outcomes in cancer. NPJ Precis Oncol 2022;6:96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. FDA . Summary of safety and effectiveness data (SSED) P180043 2019 [cited 2024 Aug]. Available from:https://www.accessdata.fda.gov/cdrh_docs/pdf18/P180043B.pdf.
- 9. FDA . Highlights of prescribing information, Balversa; 2019[cited 2024 Aug]. Available from:https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/212018s000lbl.pdf. [Google Scholar]
- 10. ADVI . Payer coverage policies of tumor biomarker and pharmacogenomic testing. 2023[cited 2024 Aug]. Available from:https://www.fightcancer.org/sites/default/files/acs_can_payer_coverage_policies_of_tumor_biomarker_and_pharmacogenomic_testing_-_advi_final.pdf.
- 11. FDA . Laboratory developed tests, FDA. 2024[cited 2024 Aug]. Available from:https://www.fda.gov/medical-devices/in-vitro-diagnostics/laboratory-developed-tests.
- 12. Zehir A, Nardi V, Konnick EQ, Lockwood CM, Long TA, Sidiropoulos N, et al. SPOT/Dx pilot reanalysis and college of American pathologists proficiency testing for KRAS and NRAS demonstrate excellent laboratory performance. Arch Pathol Lab Med 2024;148:139–48. [DOI] [PubMed] [Google Scholar]
- 13. Kim AS, Bartley AN, Bridge JA, Kamel-Reid S, Lazar AJ, Lindeman NI, et al. Comparison of laboratory-developed tests and FDA-approved assays for BRAF, EGFR, and KRAS testing. JAMA Oncol 2018;4:838–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Beaver JA, Pazdur R. The wild west of checkpoint inhibitor development. N Engl J Med 2022;386:1297–301. [DOI] [PubMed] [Google Scholar]
- 15. Ratcliffe MJ, Sharpe A, Midha A, Barker C, Scott M, Scorer P, et al. Agreement between programmed cell death ligand-1 diagnostic assays across multiple protein expression cutoffs in non–small cell lung cancer. Clin Cancer Res 2017;23:3585–91. [DOI] [PubMed] [Google Scholar]
- 16. Tsao MS, Kerr KM, Kockx M, Beasley M-B, Borczuk AC, Botling J, et al. PD-L1 immunohistochemistry comparability study in real-life clinical samples: results of blueprint phase 2 project. J Thorac Oncol 2018;13:1302–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Rugo HS, Loi S, Adams S, Schmid P, Schneeweiss A, Barrios CH, et al. PD-L1 immunohistochemistry assay comparison in atezolizumab plus nab-paclitaxel-treated advanced triple-negative breast cancer. J Natl Cancer Inst 2021;113:1733–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zajac M, Scott M, Ratcliffe M, Scorer P, Barker C, Al-Masri H, et al. Concordance among four commercially available, validated programmed cell death ligand-1 assays in urothelial carcinoma. Diagn Pathol 2019;14:99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Noske A, Wagner D-C, Schwamborn K, Foersch S, Steiger K, Kiechle M, et al. Interassay and interobserver comparability study of four programmed death-ligand 1 (PD-L1) immunohistochemistry assays in triple-negative breast cancer. Breast 2021;60:238–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Torlakovic E, Lim HJ, Adam J, Barnes P, Bigras G, Chan AWH, et al. “Interchangeability” of PD-L1 immunohistochemistry assays: a meta-analysis of diagnostic accuracy. Mod Pathol 2020;33:4–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Lawson NL, Scorer PW, Williams GH, Vandenberghe ME, Ratcliffe MJ, Barker C. Impact of decalcification, cold ischemia, and deglycosylation on performance of programmed cell death ligand-1 antibodies with different binding epitopes: comparison of 7 clones. Mod Pathol 2023;36:100220. [DOI] [PubMed] [Google Scholar]
- 22. FDA . Oncology drug products used with certain in vitro diagnostics pilot program; 2023[cited 2024 Aug]. Available from:https://www.fda.gov/medical-devices/in-vitro-diagnostics/oncology-drug-products-used-certain-in-vitro-diagnostics-pilot-program.
- 23. Friends of Cancer Research . Data Generation (and review considerations) for use of a companion diagnostic for a group of oncology therapeutic products; 2019 [cited 2024 Aug]. Available from:https://friendsofcancerresearch.org/wp-content/uploads/Panel-3_CDx_2019AM_2.pdf.
- 24. Fitzgibbons PL, Bradley LA, Fatheree LA, Alsabeh R, Fulton RS, Goldsmith JD, et al. Principles of analytic validation of immunohistochemical assays: guideline from the college of American pathologists pathology and laboratory quality center. Arch Pathol Lab Med 2014;138:1432–43. [DOI] [PubMed] [Google Scholar]
- 25. Cooper L, Chen J. Changes in companion diagnostic labelling: implementation of FDA’s april 2020 guidance for industry for in vitro CDx labeling for specific oncology therapeutic groups. Ther Innov Regul Sci 2022;56:689–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. PMDA . Guidance on drug-agnostic companion diagnostics; 2022[cited 2024 Aug]. Available from:https://www.pmda.go.jp/files/000248182.pdf.
- 27. FDA . CDRH announces intent to initiate the reclassification process for most high risk IVDs; 2024[cited 2024 Aug]. Available from:https://www.fda.gov/medical-devices/medical-devices-news-and-events/cdrh-announces-intent-initiate-reclassification-process-most-high-risk-ivds.
- 28. Jørgensen JT. Missing companion diagnostic for US food and drug administration–approved hematological and oncological drugs. JCO Precis Oncol; 2022;6:e2200100. [DOI] [PubMed] [Google Scholar]