Foundation Medicine's Regulatory Win for NGS CDx Sets Precedent; Adoption Challenge Remains

NEW YORK (GenomeWeb) – The approval of the first next-generation sequencing companion diagnostic is an important regulatory precedent given the increasing adoption of comprehensive genomic profiling in cancer care.

The US Food and Drug Administration-approved label for Foundation Medicine's FoundationFocus CDxBRCA includes information that doctors can use to learn about the strengths and weaknesses of the test. Based on the types of evidence in the label, other CDx developers can also consider if they are interested in pursuing a similar path. While Foundation Medicine gets credit for this key accomplishment, in the real world, FDA approval is still no assurance of market adoption.

The FDA this week approved FoundationFocus CDxBRCA to identify advanced ovarian cancer patients who have mutations in their BRCA1 and BRCA2 genes and are therefore more likely to benefit from Clovis Oncology's PARP inhibitor Rubraca (rucaparib). In order to garner FDA approval, Foundation had to demonstrate the test's performance in pivotal drug trials and compare it to other tests. The assay showed a 97 percent overall percent agreement with an externally developed NGS assay and 95 percent positive agreement with local BRCA tests.

Moreover, Foundation's tissue-based test can detect germline and somatic mutations (though it cannot distinguish between them), which according to the company may identify twice as many women eligible for treatment compared to blood- and saliva-based tests on the market that gauge only germline mutations. Despite the evidence underlying the diagnostic and its advantages relative to other tests, Foundation Medicine may still face an uphill battle in getting doctors to use its CDx in lieu of numerous others on the market.

During a call with analysts yesterday to discuss Rubraca's approval, Clovis CEO Patrick Mahaffy said that though Foundations' is an excellent assay, "in the real world, physicians are able to choose what tests they prefer." These include a range of lab-developed tests for gauging hereditary cancer risk and Myriad Genetics' BRACAnalysis CDx — the first laboratory-developed companion test the FDA approved two years ago alongside AstraZeneca's PARP inhibitor Lynparza (olaparib).

Mahaffy's statement underscores the longstanding difficulty for diagnostics developers that invest in taking a companion test through the FDA, only to find that effort is not rewarded in terms of market adoption since that test has to compete with other lab-developed assays that can be available without FDA approval.

This challenge is not going away any time soon, since the FDA recently said it would hold off its plan to regulate lab-developed tests, which would have required that LDTs used to make treatment decisions garner the agency's approval. 

The BRCA testing paradigm is further complicated by the fact that mutations in these genes are important for not only directing PARP inhibitor treatment, but also for gauging hereditary cancer risk. Therefore, patients may receive BRCA testing long before they develop ovarian cancer, progress on chemotherapy, and become eligible for testing with Foundation's test.

For Clovis, despite having invested in taking a CDx through the FDA, the simultaneous existence and use of LDTs serves its interests, which is to identify all the patients who might be eligible for its drug. During the call, Mahaffy acknowledged the reality that ovarian cancer patients may have their BRCA status determined by other available tests, but he added that both Clovis' sales team of 85 reps and Foundation's reps will be educating doctors about the need for tissue-based testing and importance of detecting both somatic and germline mutations. Although Foundation did not respond to questions for this article ahead of publication, the company will undoubtedly highlight the fact that Myriad's FDA-approved CDx detects only germline BRCA mutations in blood samples, and would miss ovarian cancer patients who acquire these mutations only in their tumors.

Based on market research, "the enthusiasm to have an option for the treatment of this large population of patients with somatic mutations, who have had no opportunity [for testing] before, is very high," Mahaffy told analysts during the call yesterday. "The desire to treat those patients will drive a great amount of tissue-based testing."

Test specifications

A major advantage of garnering FDA approval is that the review process results in a test label that physicians can use to understand the assay's strengths and weaknesses, which may be especially helpful in a crowded diagnostic market like the BRCA testing space.

Clovis garnered FDA approval for Rubraca for advanced ovarian cancer based on data from two single-arm studies involving approximately 100 patients, 54 percent of whom saw their tumors shrink and experienced a 9.2-month median duration of response. The company used Foundation Medicine's CDx to confirm the presence of BRCA mutations in 96 percent of study participants, and performed bridging studies to demonstrate that patients with BRCA alterations identified by the CDx or by another locally performed test had similar response rates. 

FoundationFocus CDxBRCA utilizes the Illumina HiSeq 4000 platform and sequencing is performed at more than 500x coverage, with at least 99 percent of exons sequenced at greater than 100x coverage. 

According to the test label, the performance of the CDx is not established for insertions longer than 10 nucleotides, deletions longer than 12 nucleotides, alterations residing in polyC homopolymer runs, homozygous deletions, or large rearrangements. Bridging studies showed that two samples with large rearrangements were detected by a locally performed test but not by FoundationFocus CDxBRCA.

When the CDx was compared against a clinical trial assay developed by Foundation Medicine there was 98 percent overall agreement, 97 percent positive agreement, and 100 percent negative agreement. Meanwhile, the CDx had 95 percent positive agreement with locally performed BRCA tests.

The label also details a comparison of the CDx against an unnamed NGS assay in analyzing 36 tumor BRCA-positive and 44 tumor BRCA-negative ovarian cancer samples, including insertions (1-4 nucleotides), deletions (1-12 nucleotides), and single nucleotide variants, as well as variants in homopolymer runs. One sample failed analysis by Foundation's CDx due to low coverage and six samples failed with the comparator assay due to low coverage or allele frequency. Excluding these samples, there was 97 percent concordance between the two tests, 100 percent positive agreement, and nearly 95 percent negative agreement.

"Alterations in polyT homopolymer runs may not be reliably detected," according to the CDx label. "Alterations detected at allele frequencies below the established limit of detection are not detected consistently."

Using samples with insertions ranging from 1-4 nucleotides, deletions ranging from 1-11 nucleotides, and single nucleotide variants, including variants in homopolymer runs, the test's limit of detection for alterations in non-repetitive regions or homopolymer repeats shorter than 4 nucleotides is at 6 percent mutant allele frequency. The limit of detection for representative alterations in a homopolymer region longer than 4 nucleotides is at 15.3 percent mutant allele frequency.

The CDx label also includes a table describing the criteria for classifying protein truncating and splice site alterations as deleterious. There is also a curated list of deleterious missense alterations based on the Breast Cancer Information Core database. "If a patient is positive for any of the deleterious alterations specified in the BRCA1/2 classification, the patient may be eligible for treatment with Rubraca," the CDx label states in the intended use section.

Looking beyond BRCA

At medical meetings, Mahaffy said that Foundation and Clovis will also talk about the importance of looking beyond BRCA mutations, and testing for molecular deficiencies in homologous recombination — a DNA repair mechanism that when perturbed gives rise to cancer.

Clovis and Foundation are developing a homologous recombination deficiency (HRD) signature that they hope can identify a greater proportion of patients who will respond to Rubraca. In the randomized, Phase III ARIEL3 study, for example, Rubraca's efficacy and safety will be compared to placebo across three groups: patients with germline and somatic BRCA mutations; patients with HRD including those with BRCA mutations; and all participants. 

This study, which finished enrolling earlier this year and will report results in mid-2017, will be the confirmatory study for converting the present accelerated approval for Rubraca into full approval. It will also provide evidence for prescribing the drug in the second-line maintenance setting following treatment with a platinum-based chemo.

"PARP inhibitors are great new drugs for the treatment of a specific or even a broader patient population," Mahaffy said, highlighting that the company is looking at the drug's response in a molecularly defined subset but is also hoping for a broad indication for Rubraca in the second-line maintenance setting.

Similarly, Tesaro recently said that it would pursue approval for its PARP inhibitor niraparib in the overall recurrent ovarian cancer population after a study showed that regardless of BRCA mutation or HRD status patients derived clinically meaningful benefit from the drug compared to placebo. The company used Myriad's BRACAnalysis CDx and its MyChoice HRD NGS test to stratify patients in this study. Despite Tesaro's decision to seek a broad indication for niraparib, however, Myriad is submitting its HRD test for FDA approval based on the fact that patients with a positive HRD score on niraparib saw a 9.1-month benefit in progression-free survival compared to those on placebo, while there was a 3.1-month benefit in the HRD-negative subset. 

Regardless of the potential that Rubraca may prove efficacious in an all-comer population, Mahaffy highlighted that Clovis is using a precison oncology approach, studying the drug accross cancer indications and in molecularly defined subpopulations. He highlighted, for example, the TRITON2 and TRITON3 trials where patients with metastatic, castration-resistant prostate cancer are being enrolled if they have evidence of HRD as defined by Foundation's test. Credit Suisse analyst Kennen MacKay wrote in a note to investors this week that this molecularly defined prostate cancer population could be Rubraca's largest market opportunity.