Testing the Immune Memory: How Much Anti-PD-1/PD-L1 Therapy Is Enough?

OncologyLiveVol. 17/No. 15Volume 17Issue 15

There is some evidence that relatively short treatment courses might also be appropriate for other checkpoint inhibitors approved for treating some cancers, but the data are fairly sparse.

Jeffrey S. Weber, MD, PhD

A standard course of the immunotherapy ipilimumab (Yervoy) is well defined and, to many patients with melanoma, surprisingly short: just 4 infusions given at 3-week intervals. That’s enough to provide lasting benefit to most responders and, if the benefit fades from lack of treatment, a second round of 4 infusions will often bring it back.

There is some evidence that relatively short treatment courses might also be appropriate for the other 3 checkpoint inhibitors approved for treating some cancers, but the data are fairly sparse. The relatively mild adverse effect profiles of those newer antibodies obviated any need to minimize treatment times for most patients, so trials have typically treated patients for as long as they seemed to benefit and avoided any comparisons of different durations of therapy. There is also very little published information about responsive patients who discontinue treatment, see their conditions deteriorate, and then reinitiate treatment.

This is, to some degree, a product of time. Nivolumab (Opdivo), pembrolizumab (Keytruda), and atezolizumab (Tecentriq) are still new drugs. Hundreds of additional trials are underway and many more are at various stages of planning. These studies will provide far more information about all 3 antibodies, including information about initial responders who stop and then restart treatment. At least 1 of them—a trial designed to determine whether it’s safe to stop nivolumab after 1 year—will even provide some comparative data on treatment duration. There are, however, no plans for trials that would establish optimal treatment times for each medication with any degree of certainty.

“There are certainly indications that relatively short treatments may provide as much benefit as longer treatments, at least for many patients. For example, some trial patients who respond to these medications but stop taking them because of adverse effects seem to fare about as well as patients who undergo the full duration of scheduled therapy on protocols,” said Jeffrey S. Weber, MD, PhD, a prolific immunotherapy investigator who is deputy director of the Perlmutter Cancer Center at NYU Langone Medical Center.

“Indications aren’t proof, though. A trial that definitively compared outcomes among patients receiving treatments lasting, say, 6 months, 1 year, and 2 years could require thousands of subjects. Such trials may eventually be done by cooperative groups, but there is certainly no incentive for the companies that used ongoing treatments to gain their approvals to spend untold millions proving that 6-month treatments work just as well.”

T cells that fight disease are covered with receptors that help them distinguish normal cells from foreign invaders. They also have inhibitory receptors that check T-cell action and prevent the immune response from spiraling out of control. Among the most important of these receptors are programmed death receptor-1 (PD-1) and cytotoxic T-lymphocyte antigen-4 (CTLA-4).

When T cells encounter normal human cells, PD-1 binds with the programmed death receptor-ligand-1 (PD-L1) that’s expressed on the outside of healthy human cells, and that connection signals the T cell to stand down. When T cells encounter cells that lack PD-L1, they attack and call other T cells for backup. Many tumors evade that immune response, presumably because they express some PD-L1 and thus convince T cells to accept them as normal. Both nivolumab and pembrolizumab are anti- PD-1 antibodies that encourage the immune system to attack cancer by preventing receptors on T cells from binding with ligands on tumors. Atezolizumab works on tumors and other immune cells rather than T cells, blocking the ligands to prevent them from binding with the PD-1 on the T cells. Ipilimumab works in a broadly similar way. It attaches itself to CTLA-4 on the surface of T cells and prevents CD80 or CD86 ligands on other cells (though not usually tumor cells) from binding with the CTLA-4 and inhibiting T-cell activation and proliferation. This blockade of the CTLA-4 signaling pathway can also reduce T-regulatory cell function and (it is thought) increase T-cell response to tumors and other perceived threats.

A number of clinical trials of ipilimumab have supplemented the 4-infusion course of treatment with maintenance doses, administered every 12 weeks for various durations. Most such trials gave maintenance to all ipilimumab patients, but a few of them have randomized ipilimumab patients between genuine maintenance and placebo. Investigators continue to follow outcomes in several cases, but results to date have yet to show any significant benefit to ongoing ipilimumab.

A 2015 paper in the Journal of Clinical Oncology1, for example, reported 5-year survival rates of 502 treatment-naïve melanoma patients in a phase III trial that pitted dacarbazine and placebo against dacarbazine and ipilimumab. Patients who initially received ipilimumab could sign up for maintenance, though half of them got placebo maintenance.

After 5 years, investigators found that initial treatment with ipilimumab more than doubled survival but that the only significant effect of genuine maintenance was immune-related adverse events.

Another reason that ipilimumab is typically used without maintenance is that the phase III trial that secured its initial approval from the US Food and Drug Administration (FDA) did not use maintenance doses.2

Prescribing conventions may change, however, because the phase III trial3 that led the FDA to expand ipilimumab’s indication last year did include maintenance therapy. (The initial approval had been for inoperable melanoma. The new approval endorses its use after surgery in patients who are at high risk for recurrence.)

Nearly all ipilimumab trials—whether or not they have used maintenance doses—have used a 4-infusion sequence for initial treatment. This brief burst of initial treatment was designed to minimize the very serious immune-related adverse effects that the antibody can trigger, and even 4 doses is too much for a significant percentage of patients, who must stop using ipilimumab after just 1 or 2 infusions.

Antibodies that block the PD-1/PD-L1 pathway, on the other hand, demonstrated more favorable adverse event profiles in early clinical testing. There was no need to create standardized treatment protocols early in the trial process and no need to risk losing out on any benefits by treating patients for too short a period.

“The course of treatment has varied in PD-1 and PD-L1 inhibitor trials, but many have been treated until progression. Some patients in the KEYNOTE-001 trial, for instance, have now been on pembrolizumab for 4 years to date because the design was for an indefinite treatment period,” said Edward Garon, MD, a researcher and clinical oncologist at the Ronald Reagan UCLA Medical Center and an associate professor at UCLA’s Geffen School of Medicine.

The phase II trial4 that recently led to the FDA’s initial approval of atezolizumab gave the drug on an ongoing basis to 310 patients with urothelial carcinoma. The phase I trial5 that led the FDA to first approve pembrolizumab for ipilimumabrefractory melanoma in 2014 gave the antibody until progression to 173 patients.

The FDA later expanded pembrolizumab’s indication (in December 2015), citing 2 additional studies to support its use as an initial treatment for unresectable or metastatic melanoma. The first of these was an 834-patient phase III trial6 that randomized patients between 4 doses of ipilimumab and as much as 2 years of pembrolizumab.

The second was a phase II trial in 540 patients with ipilimumab-refractory melanoma7 that pitted each investigator’s choice of chemotherapy against ongoing pembrolizumab.

In other words, all of the trials used to support existing approvals for either antibody used either open-ended or 2-year treatment periods.

The same is true for nivolumab. The FDA initially approved it in 2014 for the treatment of advanced melanoma, after receiving data on the first 120 melanoma patients in a 405-patient phase III trial8 that treated until progression. In early 2015, the agency cited results from a 272-patient trial of ongoing nivolumab use9 to approve the drug for the treatment of advanced squamous-cell non—smallcell lung cancer. (The lung cancer indication was then expanded on the strength of a 582-patient trial10 that, again, gave nivolumab until progression.) Also in 2015, a trial of ipilimumab (4 infusions) and nivolumab (until progression) led the FDA to approve the combination for BRAF wild-type melanoma.

(The indication was later expanded across BRAF types following a 945-patient trial11 that showed that 4 infusions of ipilimumab and ongoing nivolumab worked better than either treatment given as monotherapy.) The drug’s final 2 approvals to date hinge on a study of 821 patients with advanced renal cell carcinoma,12 which lasted until disease progression, and a study of 95 patients with relapsed or refractory Hodgkin lymphoma13 which lasted until progression, complete response or 2 years of treatment.

However, despite the near ubiquity of openended treatment protocols in these studies, there is reason to believe that PD-1/PD-L1 antibodies need not be taken for life the way that targeted cancer medications are typically taken by patients who respond to them.

“Targeted medications fight cancer directly. When patients stop taking them, they stop working. So barring a complete response, patients keep taking them as long as they’re effective and tolerable,” said Patrick Ott, MD, PhD, clinical director of the Center for Immuno-Oncology at Dana-Farber Cancer Institute. “Immunotherapy, on the other hand, attacks cancer indirectly—generally by teaching long-lasting T cells to attack tumors. Once the relevant cancer specific T cells are fully primed, they often have the ability to keep at it without ongoing immunotherapy. We just don’t know how long it takes to maximize immune response, and we don’t know whether ongoing treatment or reinduction as needed is the best way to maintain that response, particularly with PD-1/PD-L1 directed therapy.”

A small number of PD-1/ PD-L1 immunotherapy studies provide relatively long-term comparative outcome data on patients who stop treatment at different times, and the results suggest that shorter treatment durations may work as well as longer treatment durations, at least for some patient groups.

For example, a presentation at this year’s ASCO meeting provided 3-year overall survival data for advanced melanoma patients treated with pembrolizumab in the KEYNOTE-001 trial.14 Of the 655 patients who initially enrolled, 95 achieved a complete response at some point, and 61 of those patients decided to move from active treatment to observation.

After an average of 10 months without pembrolizumab, all but 2 of those 61 patients had maintained a complete response (and progression had yet to be confirmed in 1 of the other 2 at the time of the presentation). Most of the complete responders abandoned treatment only after more than a year (median treatment duration: 23 months), but the patient-by-patient chart shows no correlation between treatment time and outcomes. Patients who achieved complete response before stopping have fared well—even those few patients who achieved complete response early and received less than a year of total treatment. The KEYNOTE-001 presentation did not compare outcomes between the roughly equal number of patients who were still taking pembrolizumab (21%) and those who stopped due to adverse events (24%). However, another ASCO presentation did provide such information from a smaller study, the CheckMate-069 trial of nivolumab plus ipilimumab in patients with advanced melanoma.15

Of the 95 patients who were initially randomized to the combination treatment, 35 dropped out at some point due to adverse events. The overall response rate was actually higher among dropouts (66%) than among the total combination therapy cohort (59%), as was the rate of partial response (46% vs 37%) and stable disease (17% vs 13%) but not complete response (20% vs 22%). Tumor reduction burden was virtually identical. Among the 23 patients who showed some response before dropping out, 17 of them (74%) continued to respond to the end of the study period. Treatment periods for those patients typically lasted just 2 to 4 months and response durations had typically exceeded 18 months. The rate of ongoing response was higher among all 65 combination therapy patients who showed some initial response (80% vs 74%) as was their 2-year rate of overall survival (71% vs 64%). Still, the 2-year rate of progression-free survival was virtually identical (52% vs 51%), and the investigators who conducted the analysis concluded that patients who continued combination therapy and those who stopped early derived similar benefits.

Short-term immunotherapy users have not fared so well in all studies. A presentation of data from the KEYNOTE-010 study of pembrolizumab versus docetaxel for previously treated advanced non—smallcell lung cancer found that pembrolizumab was associated with significantly longer overall survival, except in those who took it for less than 18 weeks. Garon noted that this result doesn’t necessarily argue for longer treatment. Those who quit early were likely nonresponders who quit because of disease progression. Longer treatment probably would not have improved responses.16

As investigators work to determine optimal treatment times for anti-PD-1/PD-L1 therapies, they also hope to discover how frequently those drugs work when they are used for a second (or more) time on patients who have responded, discontinued treatment, and started to progress again. There is already strong evidence that the majority of patients who respond to ipilimumab once will benefit from a second course of 4 infusions.17 The evidence to date on the other 3 antibodies consists mainly of case studies, albeit case studies that typically find repeat response.18

So how long should clinicians treat patients with PD-1/PD-L1 immunotherapy while they await new data? Strategies suggested by the physicians interviewed for this story, who all rank among the top immunotherapy researchers, varied. The shortest protocol anyone recommended was to discontinue treatment after a patient’s condition stops improving and remains stable for a couple months (so long as the patient has undergone a full year of treatment). The longest protocol anyone recommended was ongoing treatment for patients who continue responding without adverse effects or financial hardship.

That said, all 4 physicians said that such a medically uncertain decision should rest, in large part, with each individual patient and that individual patient preferences tend to fall in 2 distinct camps: those who want to abandon treatment as soon as they stop improving and those who see no reason to alter treatment as long as they remain stable. “Decisions right now are guided more by patient emotions than by scientific evidence because we simply don’t have enough science to guide them,” said Michael A. Postow, MD, a researcher and clinician at Memorial Sloan Kettering Cancer Center.

“Looking forward, the good news is that we’re going to get significantly more evidence to guide decisions. Institutions will be able to analyze outcomes for patients who have dropped out of trials. Other studies are investigating defined treatment periods. “The bad news is the general lack of randomized trials designed to compare different treatment durations. We will be left to extrapolate optimal treatment times by comparing results from different studies or comparing people who leave trials to those who don’t. Those comparisons should tell us something, but they’re not robust enough to provide definitive answers,” Postow said.


  1. Maio M, Grob JJ, Aamdal S, et al. Five-year survival rates for treatment-naïve patients with advanced melanoma who received ipilimumab plus dacarbazine in a phase III trial. J Clin Oncol. 2015;33(10):1191-1196. doi:10.1200/JCO.2014.56.6018.
  2. Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363(8): 711-723. doi:10.1056/NEJMoa1003466.
  3. Eggermont A, Chiarion-Sileni V, Grob JJ, et al. Adjuvant ipilimumab versus placebo after complete resection of high-risk stage III melanoma (EORTC 18071): a randomized, double-blind, phase 3 trial. Lancet. 2015;16(5):522-530. doi: 10.1016/S1470-2045(15)70122-1.
  4. Rosenberg JE, Hoffman-Censits J, Powles T, et al. Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy; a single-arm, multicenter, phase 2 trial. Lancet. 2016;387(10031):1909-1920. doi:10.1016/S0140-6736(16)00561-4.
  5. Robert C, Ribas A, Wolchok JD, et al. Anti-programmed-death-receptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: a randomized dose-comparison cohort of a phase 1 trial. Lancet. 2014;384(9948):1109-1017. doi:10.1016/S0140-6736(14)60958-2.
  6. Robert C, Schachter J, Long GV, et al. Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J Med. 2015;372(26):2521- 2532. doi:10.1056/NEJMoa1503093.
  7. Ribas A, Puzanov I, Dummer R, et al. Pembrolizumab versus investigator- choice chemotherapy for ipilimumab-refractory melanoma (KEYNOTE-002): a randomized, controlled, phase 2 trial. Lancet. 2015;16(8):908-918. doi:10.1016/S1470-2045(15)00083-2.
  8. Weber JS, D’Angelo SP, Minor D, et al. Nivolumab versus chemotherapy in patients with advanced melanoma who progressed after anti-CTLA-4 treatment (CheckMate 037): a randomized, controlled, open-label, phase 3 trial. Lancet. 2015;16(4):375-384. doi:10.1016/S1470-2045(15)70076-8.
  9. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous-cell non—small-cell lung cancer. N Engl J Med. 2015;373(2):123-135. doi:10.1056/NEJMoa1504627.
  10. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced nonsquamous non—small-cell lung cancer. N Engl J Med. 2015;373:1627-1639. doi:10.1056/NEJMoa1507643.
  11. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373:23-24. doi:10.1056/NEJMoa1504030.
  12. Motzer RJ, Escudier B, McDermott DF, et al. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;373(19):1803-1813. doi:10.1056/NEJMoa1510665.
  13. Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin’s lymphoma. N Engl J Med. 2015;372:311-319. doi: 10.1056/NEJMoa1411087.
  14. Robert C. Three-year overall survival for patients with advanced melanoma treated with pembrolizumab in KEYNOTE-001. Presentation presented at: 2016 ASCO Annual Meeting; June 3-7, 2016; Chicago, IL.
  15. Hodi S. Overall survival in patients with advanced melanoma (MEL) who discontinued treatment with nivolumab (NIVO) plus ipilimumab (IPI) due to toxicity in a phase II trial (CheckMate 069). Poster presented at: 2016 ASCO Annual Meeting; June 3-7, 2016; Chicago, IL.
  16. Garon EB. Pembrolizumab versus docetaxel for previously treated advanced NSCLC with a PD-L1 tumor proportion (TPS) 1%-49%: results from KEYNOTE-010. Poster presented at: 2016 ASCO Annual Meeting; June 3-7, 2016; Chicago, IL.
  17. Robert C, Schadendorf D, Messina M, Hodi FS, O’Day S. Efficacy and safety of retreatment ipilimumab in patients with pretreated advanced melanoma who progressed after initially achieving disease control. Clin Cancer Res. 2013;19(8):2232-2239. doi:10.1158/1078-0432.CCR-12-3080.
  18. Lispon EJ, Sharfman WH, Drake CG, et al. Durable cancer regression off-treatment and effect re-induction therapy with an anti-PD-1 antibody. Clin Cancer Res. 2013;19(2):462-468. doi:10.1158/1078-0432. CCR-12-2625.
Related Videos
View All
Related Content