Alexander M. Castellino, PhD

June 20, 2016

CHICAGO — At present in clinical practice, immunotherapy with anti-PD-1 agents is administered indefinitely until intolerable toxicities or progressive disease sets in. But there has been anecdotal evidence that patients who stop treatment may still derive benefit, and now there is evidence of this from a post hoc analysis of a randomized study.

It comes from the CheckMate 069 trial that evaluated the combination of nivolumab (Opdivo, Bristol-Myers Squibb Company) and ipilimumab (Yervoy, Bristol-Myers Squibb Company) vs ipilimumab alone in patients with metastatic melanoma.

New results from a post hoc analysis of this trial, presented at the recent American Society of Clinical Oncology (ASCO) 2016 Annual Meeting (abstract 9518), show that a subgroup of patients who discontinued combination immunotherapy because of treatment-related adverse events achieved an impressive overall response rate (ORR) of 66%.

"This study is descriptive about what happens to patients when they discontinue treatment with the combination of immunotherapy," lead author Michael A. Postow, MD, a medical oncologist at Memorial Sloan Kettering Cancer Center, New York City, told Medscape Medical News. "So many patients have to discontinue because of the side effects of the combination. That's why it is important to know what happens to these patients," he added.

High Rate of Discontinuation

There was a high rate of discontinuation of the combination immunotherapy regimen (nivolumab plus ipilimumab) in both the phase 2 Checkmate 069 trial and the later phase 3 Checkmate 067 trial that led to the approval of the combination.

In both studies, the combination was associated with grade 3/4 adverse events, which occurred in more than half the patients (CheckMate 069, 55%; CheckMate 067, 54%). More than a third of patients stopped treatment because of these events (CheckMate 069, 36.4%; CheckMate 067, 37%).

Patients in the combination arm received nivolumab 1 mg/kg and ipilimumab 3 mg/kg every 3 weeks for four cycles and then nivolumab 3 mg/kg until disease progression or unacceptable toxicity.

In the Checkmate 069 trial, the baseline characteristics for patients who discontinued treatment (n = 35) were generally similar to those of patients who were initially assigned to receive the combination (n = 94). Of the 35 patients who discontinued treatment, 77% (27/35) did so during the combination phase of treatment. Patients received three median doses of nivolumab (range, 1-45) and three median doses of ipilimumab (range, 1-4)

"This minimizes the criticism that discontinuation may have occurred after responses were achieved," Dr Postow said. "The majority of patients discontinued before the response was known," he added.

The ORR seen in patients discontinuing treatment was high — 66% (complete response [CR], 20%; partial response [PR], 46%; stable disease [SD], 17%).

These responses were similar to those reported for the overall population randomly assigned to the combination (ORR, 59%; CR, 22%; PR, 37%; SD, 13%).

Disease progression occurred in 16% of all patients and in 9% of patients who discontinued therapy. Rates of reduction of tumor burden were also similar (-69% for patients who discontinued and -70% for all patients).

At 2-year follow-up, median overall survival (OS) and progression-free survival (PFS) were not reached for patients who discontinued therapy — an observation similar to that noted for the randomized population. Two-year OS and PFS rates were 71% and 52%, respectively, for patients who discontinued therapy; corresponding rates in patients assigned to the combination were 64% and 51%, respectively.

Why did patients discontinue treatment? Most did so because of grade 3 adverse events (25%: colitis, diarrhea, elevated liver enzyme levels, pneumonitis, and enterocolitis, among others). Discontinuation for grade 2 and grade 4 adverse events occurred in 6% and 5% of patients, respectively.

In patients who discontinued therapy because of adverse events, resolution of the grade 3/4 events occurred in 92% (23/25) of patients, compared with 87% (33/38) for patients who received the combination.

"In this post hoc analysis, patients who experience treatment-related adverse events leading to discontinuation appeared to derive similar benefit from the combination despite discontinuing therapy early," the study researchers noted.

Corroboration of Data From CheckMate 067

The findings reported by Dr Postow are similar to those seen in the phase 3 CheckMate 067 study, which was published last year (N Engl J Med. 2015;373:23-34).

In answer to a letter to the editor, the CheckMate 067 researchers note, "Among the 120 patients who discontinued combination therapy because of toxic effects, the response rate was 67.5%."

"To us, this supports the current guidelines for management of toxic effects and discontinuation of treatment, since it shows that high response rates can be observed in the context of no treatment-related deaths. Longer follow-up will be needed to assess the effect of treatment discontinuation on overall survival," they add.

Do the Findings Inform Clinical Practice?

As yet, there are no results of randomized trials informing physicians as to the optimum length of immunotherapy. The question is an important one, not least because of the high cost of these new drugs.

At the ASCO meeting, discussant Marc S. Ernstoff, MD, of Roswell Park Cancer Center, Buffalo, New York, provided an insight into annual drug cost based on wholesale acquisition drug prices. At approved dosages and schedules, for a 70-kg adult, 1 year of immunotherapy was estimated to cost $116,532 for pembrolizumab (Keytruda, Merck Sharp & Dohme Corp), $130,931 for nivolumab, $146,160 for ipilimumab, and $286,202 for the combination of nivolumab and ipilimumab.

Dr Postow explained that in clinical practice, treatment would be discontinued for patients experiencing significant immune-related adverse events. However, for patients who do not experience intolerable side effects, the question is open as to when physicians may choose to discontinue treatment. "We are missing long-term follow-up to advise patients on stopping treatment if they experienced a response," he said. "The depth and duration of response may be important," he added.

At the ASCO meeting, a clue to this question was provided in a presentation of long-term follow-up data from the KEYNOTE-001 trial for patients receiving pembrolizumab. Caroline Robert, MD, PhD, of Gustave Roussy and Paris-Sud University, France, reported a 3-year OS rate of 40%, a median OS rate of 24.4 months, and a long-term ORR of 33%, as previously reported by Medscape Medical News.

Dr Robert reported that responses were maintained in nearly all patients (59 of 61 patients, 97%) who stopped treatment after achieving a CR and were under observation. Only two of these patients experienced disease progression after discontinuance of treatment. For these patients, the median amount of time on treatment was 24 months; median period during which they were off treatment was 10 months. The median time to first response was 3 months, and the median time to CR was 13 months. Median duration of response was not reached. "Complete responses are durable," Dr Robert said.

"From these data, I feel a bit more comfortable with stopping treatment once a CR has been reached," Dr Postow said. He explained that responses with immunotherapies can change over time — stable disease can convert to PR, and PR can convert to CR.

Dr Postow also explained that patients might discontinue and restart therapy, an approach that has the potential for excellent benefits. This was seen with nivolumab from a phase 1 study, as reported at the American Association of Cancer Research 2016 meeting, After 5 years, 34% of 107 patients were still alive after 5 years, as previously reported by Medscape Medical News. In five of these patients, treatment was stopped and was subsequently restarted after the patients experienced disease progression.

But apart from these few details from clinical trials and a few case reports, there are not much data on discontinuing immunotherapy that can be used to guide clinical practice, Dr Postow summarized.

Several studies are in planning stages, but protocols still need to be finalized, according to Dr Postow. In the meantime, physicians will have to use their best clinical judgement and discuss with their patients the risks and benefits of discontinuing therapy, he said.

The Checkmate studies were funded by Bristol-Myers Squibb. Dr Postow has played a consulting or advisory role with Amgen and Bristol-Myers Squibb. He has received research funding, as well as travel and accommodation expenses, from Bristol-Myers Squibb.

American Society of Clinical Oncology (ASCO) 2016 Annual Meeting: Abstract 9518


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.