Cancer specialists like me, whether they are medical, surgical, or radiation oncologists spend our lives treating life threatening disease. We generally have a serious and professional demeanor reflective of the world that we choose to inhabit and the circumstances facing our patients that entrust us with their care.
Employees of FDA are charged with regulatory authority regarding drugs used by oncologists and other medical practitioners. They understand that their task requires them to make dispassionate judgments based upon a record of historical precedent, expert advice, and validated endpoints to both protect patients and facilitate their access to new discoveries. Those FDA employees are generally known for their equanimity and diplomacy.
On Feb. 9, FDA convened the Oncologic Drug Advisory Committee to hold an unprecedented hearing about how to manage a potential paradigm-busting development in the management of a subset of patients with stages 2 and 3 locally advanced rectal cancer (LARC). The preliminary data that the oncologists and their regulators reflected on is enough to make oncologists giddy and FDA regulators bold.
They assembled to hear evidence and commentary and then get an up or down vote from ODAC on whether the path proposed by GlaxoSmithKline for evaluating the safety and efficacy of their PD-1 inhibitor, dostarlimab, could lead to a new indication as an initial treatment for LARC.
GSK is endeavoring to gain approval for their drug as an alternative to standard of care multimodality therapy in the subgroup of patients with LARC whose tumors exhibit microsatellite Instability (MSI). Such tumors are classified as MSI-High (MSI-H).
Patients with MSI-H tumors and advanced disease, including colorectal cancer patients, have enjoyed remarkable benefit from treatment with a panoply of PD-1 and PD-L1 inhibitors.
MSI-H tumors have a defect in DNA mismatch repair mechanisms that generate multiple tumor-induced cell surface neoantigens that provide a target for T cells once the T cells are engaged using PD-1 or PD-L1 inhibitors.
Some patients who received these drugs for treatment of refractory advanced disease, remarkably, remain disease-free years later. To borrow a phrase that I heard long ago from my fellow medical residents, “they beat the reaper.” We, oncologists, celebrate when our patients beat the reaper.
PD-1 inhibitors have already broken the mold with respect to FDA approvals. Pembrolizumab, developed by Merck, received an unprecedented indication for treatment of patients with MSI-H tumors regardless of their tumor’s organ of origin based on efficacy across tumor types. These findings and others that followed stoked the immune-oncology revolution that is changing practices and leading to outcomes that were recently only pipe dreams.
There are now multiple PD-1 and PD-L1 inhibitors approved for use in the United States and worldwide—and there are more in pipelines. Dostarlimab is already on the market. It is indicated for the treatment of adult patients with mismatch repair deficient recurrent or advanced endometrial cancer, as determined by an FDA-approved test, that has progressed on or following prior treatment with a platinum-containing regimen.
This indication is approved under accelerated approval based on tumor response rate and durability of response, but on Feb. 9, the day of ODAC meeting, this indication was converted to a regular approval.
Rectal cancer is all too common, with nearly 45,000 cases last year in the U.S. and 70% of rectal cancer patients have LARC. Between two and 20% of patients with stages 2 and 3 cancer have tumors that manifest MSI. The current standard of care includes triple modality therapy with 5-FU based chemotherapy, radiation therapy, and surgery regardless of MSI status. There are data that suggest that patients with LARC that is MSI-H are less responsive to chemotherapy than patients whose tumors lack that finding.
To borrow a phrase that I heard long ago from my fellow medical residents, ‘they beat the reaper.’ We, oncologists, celebrate when our patients beat the reaper.
Recent clinical trials have engendered refinements in LARC management such that total neoadjuvant therapy, where chemotherapy and radiation therapy precede surgery, have modestly improved survival outcomes while improving patient quality of life compared to older approaches where surgery preceded other modalities. These treatments are long, arduous, expensive, and have the potential for long-term toxicity.
Some patients die from treatment related side effects during this therapy. These toxicities include the need for a temporary or permanent ostomy, rectal dysfunction leading to soiling and discomfort, infertility, sexual dysfunction, peripheral sensory neuropathy and other outcomes that makes patients trade hope of cure for lifelong sequelae.
Progress has been incremental. Many oncologists and most patients are impatient for the field to make bigger forward strides. Some patients, especially those who develop recurrent disease after adjuvant therapy, favor following the lead of AIDS activists in raising their voices for investments that lead to more than incremental gains (The Cancer Letter, July 15, 2022).
Recently, as therapies have become more effective, centers of special expertise have pioneered a strategy of omitting or deferring surgery in LARC patients who manifest clinical complete remissions (cCRs) after neoadjuvant therapy. Such a response is a biological marker that portends better survival outcomes than are observed in the cohorts of patients with incomplete responses or progressive disease.
This approach is termed “Watch and Wait” (W&W) and the data supporting it comes from nonrandomized and largely single institution series. W&W demands attentive follow-up from the oncology team and compliance on the part of the patient to permit expert and potentially curative intervention if locally recurrent disease develops.
Patients who are managed with W&W can be cured despite local recurrence with “salvage surgery” and do not appear to be subject to higher rates of distant recurrence than those managed with trimodality therapy. There have been no randomized trials comparing W&W to trimodality therapy to date.
Because this approach involves no new drugs or devices, its deployment is a matter of agreement by multidisciplinary teams rather than requiring permission granted by regulatory agencies. Despite the lack of phase III trial data, W&W is an increasingly commonly deployed management strategy—and patients welcome it.
There has also been an interest in determining whether radiation can be safely withheld in the treatment of some patients and the results of the NCI-funded PROSPECT trial testing this strategy are expected later this year. That large phase III trial has taken years to accrue and more years for the results to mature.
Depending on its results, we may find that we have exposed patients who did not need radiation to its long-term toxicities. Radiation toxicity, although its frequency and severity has been moderated using newer techniques, often results in infertility, sexual dysfunction, and bowel-related issues.
The gold standard of large, randomized trials with 7-10 year times to maturity often means that the field has evolved past the question that the trial poses by the time the answer is clear. We hope that will not be the outcome once the PROSPECT data are revealed.
There are data from the NICHE and NICHE-2 studies where patients with stages 2 and 3 colon cancer were treated with nivolumab (another PD-1 inhibitor) or nivolumab plus the CTLA-1 inhibitor ipilumimab prior to surgery. A high number of responses and cCRs were documented. Since the adjuvant therapy of colon cancer does not require radiation and is managed with the less complicated surgery its treatment leads to fewer long term side effects.
The potential impact of this approach, while substantial, is less likely to generate the marked improvement in long-term quality of life that could come from eliminating trimodality therapy in LARC.
There is a large, NCI funded phase III study for patients with MSI-H stage III colon cancer that has completed accrual and is maturing. These patients were randomized after surgical resection to a fluoropyrimidine with oxaliplatin plus or minus atezolizumab, another PD-L1 inhibitor developed by Genentech. It is likely that it will take this study years before its results are unblinded and made public.
These considerations paved the way for the phase II study that broke the paradigm in rectal cancer and engendered this ODAC meeting. It is a GSK-sponsored, investigator-initiated study conducted by Andrea Cercek, Luis Diaz, and their colleagues at Memorial Sloan Kettering Cancer Center.
The interim results of this ongoing study were striking, leading to their concurrent presentation at ASCO’s 2022 annual meeting and publication in the New England Journal of Medicine. Patients enrolled in the trial received dostarlimab IV every three weeks for 9 cycles, and then underwent meticulous restaging, using endoscopy and imaging.
The study had as a primary endpoint the cCR at 12 months. At the time of the report 12/16 enrolled patients had been observed for 12 months. All 16 patients had a cCR. Total enrollment of 30 patients was rapidly achieved after the preliminary data became public and follow up continues. cCR is a novel endpoint that lacks validation in randomized trials.
The report was greeted with similar enthusiasm to that seen when the pembrolizumab data were reported in MSI-H tumors. One benefit to this study design is the fact that it was an open-label, single-arm trial, so the outstanding activity was not hidden to the investigators behind a firewall, as is the case in phase III trials.
GSK has proposed to FDA the completion of three clinical trials to support a new indication for dostarlimib for treatment of patients with MSI-H LARC. The first study is the single-institution Cercek trial reported above, which has met its accrual goals.
They propose a second multi-institutional and international study of 100 patients with stage II and III LARC using the same approach as pursued by Cercek et al. Patients with cCR would enter a W&W strategy with additional therapy only if they manifested recurrent disease.
This is a departure from FDA standards that have heretofore demanded randomized phase III trials for efficacy in the setting of adjuvant therapy undertaken with curative intent. The sponsors have also proposed doing a third randomized phase III trial of dostarlimab plus surgery versus standard adjuvant therapy followed by surgery in patients with stage 3 colon cancer.
In developing their plan, GSK solicited advice from patients who are members of its patient advisory group as well as other patients who have been treated for LARC. These patients were adamant that a randomization to standard of care versus dostarlimab would not accrue patients who would be unwilling to accept trimodality therapy as an alternative.
So, FDA is left with a dilemma. Do they demand that a drug that shows so much early promise for erasing the long-term compromises that patients must accept be tested in a trial that may never accrue patients?
Since the drug is approved for another indication, there is a real likelihood that well-resourced patients will be able to obtain it while less well-resourced patients will face unsurmountable barriers to getting that treatment.
If you had MSI-H LARC wouldn’t you want it? I would.
One would think that most federally convened meetings to dissect data would be predictably routine. They are not. ODAC meetings regularly produce great drama, and this one did not disappoint on that score.
The ODAC hearing featured the testimony of innovators who have generated these unprecedented data in patients with MSI-H LARC, experts unconnected to the trial who put these findings in perspective, FDA officials who presented issues that arise from these unexpected findings, and representatives from the drug’s sponsor, GSK.
The FDA asked the members of ODAC to evaluate the following five items:
Task 1: Evaluate the adequacy and appropriateness of the proposed single-arm trials to evaluate the efficacy and safety of dostarlimab, including the long-term benefits and risks of treatment in the indicated population.
Outcome: The consensus after discussion was that, despite reservations on the part of some ODAC members, permitting this approach is a necessary and pragmatic compromise to evaluate a potentially practice changing approach.
Task 2: Evaluate the adequacy of the proposed clinical endpoints (clinical complete response rate, event free survival), to characterize and verify the benefit of dostarlimab, including the proposed timing of the analyses of these endpoints.
Outcome: The consensus after discussion was that despite reservations on the part of some ODAC members permitting cCR at 12 months to be the primary endpoint is a necessary and pragmatic compromise to evaluate a potentially practice changing approach. Clearly the committee members wanted longer follow-up including survival data and want to hold GSK to a high standard of data sharing and data analysis.
Task 3: Evaluate the appropriateness of the study population comprising patients with stage 2/3 LARC dMMR/MSI-H for a non-operative treatment approach.
Outcome: The prevailing consensus after discussion was that stage II and III LARC patients is the appropriate study population.
Task 4: What is the appropriate role of data from a study evaluating dostarlimab for the treatment of locally advanced colon cancer to provide supportive evidence of dostarlimab’s safety and effectiveness as a treatment for dMMR/MSI-H LARC.
Outcome: The prevailing consensus after discussion was that additional data on safety and efficacy would be valuable although there were reservations voiced about this study’s design and whether it will appeal to potential enrollees.
Task 5: What is the potential impact of the variability in care, expertise, etc., across multi disciplinary study staff and across study sites on study conduct and ultimately, on clinical outcomes?
Outcome: The prevailing consensus after discussion was that GSK has devised an extensive educational program to try to minimize the variability with respect to study sites. Despite that program, site-related variability remains a concern that is beyond the scope of the hearing and it should not jeopardize evaluating this new therapy.
After nearly six hours of presentations and discussions, the ODAC members were asked to vote and to present their justifications for their vote. The committee voted to permit the plan brought forward by GSK for a non-randomized international multicenter phase II study with an accrual goal of 100 patients to proceed.
It was a yes; eight in favor, five opposed. The opposition mainly focused around the validity of cCR at 12 months as a primary endpoint.
It was notable that the four GI oncologists in the committee all voted in favor of the approach. The vote only approves the blueprint, and the approval of the drug for this indication awaits accrual and follow-up and analysis.
FDA is to be commended for putting together a comprehensive review of the data, a panel of experts to consider the data, and for finding a way to move forward on what could be one of the most exciting developments in GI oncology in years.
Under Rick Pazdur’s stewardship, the FDA Oncology Center of Excellence has found ways to accelerate the pace of progress. Since Pazdur is a GI oncologist who has treated many patients with LARC, I am sure this issue was of special interest to him.
This recommendation by ODAC is a pragmatic compromise solution with a reasonable timeline for getting results. If the larger trial confirms the preliminary findings, not only will the results make oncologists dance for joy, but it may also permit some of our patients with LARC to dance without worrying about soiling themselves.
That will be more than incremental progress.
