Beyond Tumor Type: New Milestones on a Long Road

Luis A. Diaz, MD

After years of efforts to steer clinical trials toward molecularly defined therapeutic targets, oncology researchers recently notched 2 key successes in changing the paradigm of how patients with advanced cancers are diagnosed and treated:

On May 23, the FDA approved pembrolizumab (Keytruda) for the treatment of patients with unresectable or metastatic microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) solid tumors, marking the first time a drug gained an indication based on its efficacy against a molecular target rather than on the primary tumor site.¹

On June 3, investigators reported that 78% of patients with TRK fusion—positive cancers responded to treatment with larotrectinib (LOXO-101), potentially making the drug the first targeted therapy developed in a “tumor agnostic” manner.²

Oncology experts have described these developments as important milestones on the road to personalized medicine. At the same time, the complexities involved with translating such molecular breakthroughs into broader research protocols and clinical practice have quickly become a matter of discussion.

Although most somatic genetic aberrations can be found in multiple tumor types, preclinical models are not adequately constructed to reflect sensitivity to potential new agents in tumor microenvironments across histologies, Flaherty and colleagues cautioned in an article in the 2017 ASCO Educational Book.³

Moreover, laboratory experience has shown that primary tumors may respond differently to a targeted therapy even if they share a molecular defect, the researchers noted. “Every drug presents unique circumstances in regard to the population of patients who might benefit from it,” they wrote. “Furthermore, development agnostic of tumor type could actually slow drug development if there are differential effects across tumor types by diverting resources from enrolling patients in a predominant population or in the tumor type most like to respond.”

A notable example of disparate results from targeting the same mutation in multiple tumor types is the BRAF mutation. “The BRAF inhibitor works in 100% of the histiocytosis patients who have BRAF mutations, but it only works in about 60% of the melanoma patients, and about 40% of lung cancer patients, and it never works alone in colorectal cancers,” said David B. Solit, MD, in an in interview with OncologyLive^®.

“So it’s the same drug and the same exact BRAF V600E mutation, but that drug works differentially depending upon what cancer type you have because each of those cancers has co-mutation patterns and other lineage-specific factors that are true because of the development of those cell contexts,” said Solit, who is the Geoffrey Beene Chair and director of the Marie-Josée and Henry R. Kravis Center for Molecular Oncology at Memorial Sloan Kettering (MSK) Cancer Center in New York City.

Nevertheless, Solit said he believes the evidence for the utility of targeted therapies across tumor types has made broad genomic sequencing of patients’ tumors a necessity. “The pembrolizumab data in particular with the MSI status makes genomic analysis of every patient’s tumor now important to do just from a standard-of-care perspective,” he said. “If you don’t do it, you’re going to miss something that that patient could have responded to.”

The major barrier to conducting such testing for every patient is that payers are balking at covering the cost of next-generation sequencing (NGS), as opposed to single-gene testing, without an approved targeted agent linked to the assays, according to Solit and other experts.

Gordon B. Mills, MD, PhD, a pioneer in the field of molecularly targeted anticancer therapies, said a “culture shift” is necessary to build support for covering such testing.

“I think we’ve known for quite a while that many drugs work beyond the original tumor that they were identified in,” said Mills, who is chair of the Department of Systems Biology and head of the Kleberg Center for Molecular Markers at The University of Texas MD Anderson Cancer Center in Houston, in an interview. “The question is whether we’re going to able to leverage the MSI approval and the [larotrectinib data], and some of the other ones that will come down the pipeline soon across diseases, to justify broad testing of all patients, or at least all patients likely to benefit.”

Mills said sequencing can cost $2000 to $3000 per patient at MD Anderson, but the reimbursement from CMS can be as low as $50. “How does one balance the need to do the testing to acquire the information to say that this really does benefit patients and should be paid for versus the pushback from industry and CMS saying until you know exactly how to use it we find it is far too expensive and not sustainable?”

From a research standpoint, Mills sees the MSI approval of pembrolizumab and the molecularly focused development of larotrectinib as © ©ktsdesign/fotolia.com “incredibly exciting.”

Many Building Blocks for MSI Approval

“The key question over the next 6 months to a year is, are we going to be able to leverage them to help the breadth of patients that we should be able to help with these technologies?” Mills said.The FDA’s recognition of dMMR as the first predictive biomarker for a targeted therapy across tumor types is built upon a “strong biologic rationale, availability of commercially used diagnostics for patient identification, and the urgent, unmet medical needs of patients with refractory cancers,” Flaherty et al contended.³

Specifically, the FDA has granted an accelerated approval to pembrolizumab for the treatment of adult and pediatric patients with unresectable or metastatic MSI-H or dMMR solid tumors that have progressed after prior treatment and who have no satisfactory alternative treatment options, as well as for patients with MSI-H or dMMR colorectal cancer (CRC) following progression on a fluoropyrimidine, oxaliplatin, and irinotecan.

The approval for the PD-1—targeting immunotherapy agent was based on data from 149 patients with MSI-H or dMMR cancers enrolled across 5 single-arm clinical trials. Ninety patients had CRC and the remaining 59 had 1 of 14 other tumor types (Figure).

Overall, the objective response rate (ORR) with pembrolizumab was 39.6% (95% CI, 31.7%-47.9%), including 11 (7.4%) complete responses (CRs) and 48 (32.2%) partial responses (PRs). The ORR was 36% in patients with CRC; it ranged from 27% to 83% in patients with other tumor types (Table, page 19; Figure). The median duration of response for the entire population was not yet reached (range, 1.6 -22.7 months). Among patients who responded to pembrolizumab, 78% had responses that lasted for at least 6 months.

“This is an important first for the cancer community,” said Richard Pazdur, MD, acting director of the Office of Hematology and Oncology Products in the FDA’s Center for Drug Evaluation and Research and director of the FDA’s Oncology Center of Excellence, in a statement announcing the approval. “Until now, the FDA has approved cancer treatments based on where in the body the cancer started—for example, lung or breast cancers. We have now approved a drug based on a tumor’s biomarker without regard to the tumor’s original location.”

The pivotal data for the approval included patients from the KEYNOTE-016 (n = 58), KEYNOTE-164 (n = 61), KEYNOTE-012 (n = 6), KEYNOTE-028 (n = 5), and KEYNOTE-158 (n = 19) trials. Pembrolizumab was administered at 200 mg every 3 weeks or 10 mg/kg every 2 weeks until disease progression, unacceptable toxicity, or a maximum of 24 months.

The median age among the 149 patients was 55 years, with 36% aged 65 or older. Across the population, 77% of patients were white, 56% were male, 36% had an ECOG performance status (PS) of 0, and 64% had an ECOG PS of 1.

Two percent of patients had locally advanced, unresectable disease, and 98% of patients had metastatic disease. Among patients with metastatic or unresectable disease, the median number of prior therapies was 2. In patients with metastatic CRC, 84% had received at least 2 prior lines of therapy compared with 53% in patients with other solid tumors.

Among the majority (n = 135) of the 149 patients, MSI-H or dMMR tumor status was determined prospectively with immunohistochemistry tests for dMMR or laboratory-developed, investigational polymerase chain reaction (PCR) tests for MSI-H status. For the remaining 14 patients, MSI-H status was determined through a retrospective evaluation of 415 tumor samples using a central laboratory-developed PCR test.

In its statement on the approval, the FDA listed common adverse events (AEs) of pembrolizumab, including fatigue, pruritus, diarrhea, decreased appetite, rash, pyrexia, cough, dyspnea, musculoskeletal pain, constipation, and nausea. Immune-mediated AEs associated with pembrolizumab include pneumonitis, colitis, hepatitis, endocrinopathies, and nephritis.⁴

The FDA also noted that the label for pembrolizumab includes a “Limitation of Use” indicating that the efficacy and safety of pembrolizumab have not been established in pediatric patients with MSI-H central nervous system cancers. The accelerated approval for pembrolizumab in this setting is contingent on the results of a confirmatory trial. The approval was preceded by a breakthrough therapy designation the FDA granted to pembrolizumab in November 2015 as a treatment for patients with MSI-H metastatic CRC.

“The main data here are showing that in CRC and non-CRC, we’re seeing response rates with pembrolizumab that are quite impressive,” Luis A. Diaz Jr, MD, whose research helped paved the way for the pembrolizumab approval, said in an interview. “The progression-free survival and overall survival are quite durable, and it’s recapitulated in entirely different studies. I believe that is what led to the FDA approval.

“What’s impressive about the FDA approval is that it is not only across all tumor types, but it is in adults and children,” added Diaz, head of the Division of Solid Tumor Oncology at MSK. “It’s the first time that we’re no longer looking at tumors by the site of origin, but rather by the genetic lesion. This is incredibly exciting and is a historic approval by the FDA. In my opinion, it will knock down the walls of how we think about cancer in the future.”

Larotrectinib Results Impressive

Now, he added, “the challenge is to get everyone tested. I think that is the first step.”Findings for larotrectinib were among the clinical trial results that generated the most excitement at the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting in June. Three of 4 patients representing a range of TRK fusion—positive solid tumors responded to the novel pan-TRK inhibitor larotrectinib and remain on treatment, setting the stage for the well-tolerated oral agent to become a standard of care for adults and children with any advanced tumor harboring this mutation.

“You would be hard-pressed to find a targeted therapy, even within a single disease context, that has results like this,” said David Hyman, MD, chief of early drug development at MSK, who reported the findings during a press conference.

Larotrectinib is currently the only selective pan-TRK inhibitor in clinical trials, he pointed out. The oral small molecule is being developed by Loxo Oncology Inc under a breakthrough therapy designation that the FDA granted in July 2016.

Hyman said that larotrectinib also may be “the first cancer therapy ever developed simultaneously in both adults and children.” The trial data presented at ASCO involved 12 pediatric and 43 adult patients ranging in age from 4 months to 76 years.

TRK fusions occur when 1 of the TRK genes becomes abnormally connected to another gene. This fusion event causes the TRK gene to be turned on and the cancer to grow. Although it is estimated that TRK fusion mutations could be found in up to 5000 patients each year, Hyman stressed that “this number may significantly underestimate its prevalence due to inadequate ascertainment by many of the existing profiling technologies.”

ASCO reported that TRK fusions were first discovered in colon cancer in 1982, but only recently has NGS enabled them to be systematically detected. To date, more than 50 different partner genes that fuse with 1 of 3 TRK genes (NTRK 1, 2, and 3) have been identified.

Hyman explained that TRK fusions typically follow 2 patterns: in adults, they occur rarely in more common cancers, such as brain, lung, colon, pancreatic, sarcoma, and melanoma; or, they occur more frequently in rare cancers, such as salivary and secretory breast cancer. This pattern also prevails in the pediatric setting, where the mutation is rarely seen in glioma and sarcoma, but more frequently in infantile fibrosarcoma and congenital nephroma.

Although relatively rare, Hyman said that it is hard to find a tumor type where TRK fusions have not been reported. “One of the defining features of TRK fusions is that they are not just found in one cancer type, but can be found in dozens of different cancer types—and in both children and adults.”

The study Hyman reported on at ASCO involved 55 TRK fusion—positive adult and pediatric patients with advanced solid tumors representing 17 unique cancer types who were enrolled in 1 of 3 clinical phase I/II clinical trials. Hyman emphasized that the patients were identified on the basis of local testing, and thus provide an example of “real world” experience: “We did not perform central screening to find the TRK fusions; in fact, 15 different laboratories identified the 55 patients who are described in this study.”

Hyman reported the ORR based on investigator assessment for 50 patients for whom confirmatory response data were available; larotrectinib (100 mg) was taken twice daily. The ORR was 76% (95% CI, 62%-87%), and this includes a PR in 64% of patients and a CR in 12%. Five additional patients, who joined the study too early to have had their confirmatory scans, all had at least a PR to larotrectinib, Hyman noted. He said most of the patients classified as having a PR “were not just meeting criteria for partial response, but had very deep tumor regressions.”

Echoing other oncology experts, Hyman said the larotrectinib data illustrate the benefits of broad mutation testing. “These findings embody the original promise of precision oncology: treating a patient based on the type of mutation, regardless of where the cancer originated,” Hyman said.

“We believe that the dramatic response of tumors with TRK fusions to larotrectinib supports widespread genetic testing in patients with advanced cancer to see if they have this abnormality.”

References

1. Keytruda [prescribing information]. Whitehouse Station, NJ: Merck & Co, Inc; 2017
2. Hyman DM, Laetsch TW, Kummar S, et al. The efficacy of larotrectinib (LOXO-101), a selective tropomyosin receptor kinase (TRK) inhibitor, in adult and pediatric TRK fusion cancers [ASCO abstract LBA2501]. J Clin Oncol. 2017;35(suppl). abstracts.asco. org/199/AbstView_199_195112.html.
3. Flaherty KT, Le DT, Lemery S. Tumor-agnostic drug development. Am Soc Clin Oncol Educ Book. 2017;37:222-230. doi: 10.14694/ EDBK_173855.
4. FDA approves first cancer treatment for any solid tumor with a specific genetic feature [news release]. Silver Spring, MD: US Food and Drug Administration; May 23, 2017. www.fda.gov/ NewsEvents/Newsroom/PressAnnouncements/ucm560167.htm. Updated May 24, 2017. Accessed July 10, 2017.