Meredith Regan, ScD, and David McDermott, MD

Dr Regan is Associate Professor, Division of Biostatistics, Dana-Farber Cancer Institute; Associate Professor of Medicine, Harvard Medical School

Dr McDermott is Chief, Medical Oncology, Cancer Center and Rosenberg Clinical Cancer Center, Co-Director Immunotherapy Institute, and Director, Cutaneous and Immuno-Oncology Programs, Beth Israel Deaconess Medical Center; Leader, Kidney Cancer Program, Dana-Farber/Harvard Cancer Center; Professor of Medicine, Harvard Medical School

In the late 1990s, high-dose interleukin 2 (HD-IL2) was approved by the FDA for the treatment of metastatic melanoma, based upon long-term responses in a subset of patients in a series of single-arm clinical trials, without showing a significant improvement in overall survival in a randomized trial.1

In contrast to chemotherapy, which required persistent administration to prevent disease progression, HD-IL2 established proof of the concept that systemic immunotherapy could produce a durable response accompanied by a treatment-free interval in a small subset of patients with metastatic melanoma.2 The low response rates and significant toxicities associated with HD-IL2 limited its use to selected centers, and more tolerable and effective therapies were needed, as the median overall survival was of only 6 to 9 months.1

The introduction of single-agent and combination immune checkpoint inhibitor (ICI) therapy for patients with advanced melanoma has contributed to an improvement in overall survival to more than 4 years.3,4 Just as seen with HD-IL2, patients discontinuing ICIs may experience periods of remission or durable disease control without the need for subsequent systemic therapy.5 In patients with advanced melanoma treated with combination nivolumab plus ipilimumab in the randomized, double-blind, phase III CheckMate 067 and phase II CheckMate 069 trials,3,4 a subset of patients who discontinued nivolumab plus ipilimumab early because of treatment-related adverse events (AEs) experienced durable responses, with a median time to subsequent systemic therapy greater than 2 years.4

We propose that the analysis of treatment-free intervals—defined as the period between the end of initial therapy until need for next-line therapy—should be part of future reporting of clinical trials involving immuno-oncology agents in comparison with one another, as well as in comparison with chemotherapeutic and targeted therapies. However, the limited description of treatment-free intervals in the literature so far has typically summarized a highly-selected subset of the patients, and therefore provides a partial picture of patients’ experiences.

Therefore, we aimed to improve upon this by developing an outcome measure—defined based upon all patients—to characterize the time free of subsequent systemic therapy that may be achieved with ICIs. Plus, we incorporated the possibility of persistent and/or late side effects of initial therapy within the treatment-free interval, to more completely describe all patients’ experiences.

We are developing a novel clinical trial outcome measure, treatment-free survival (TFS), that can simultaneously characterize the antitumor activity and be partitioned to include toxicity experienced during the period after cessation of ICI protocol therapy until initiation of subsequent systemic therapy or death. We illustrate the outcome, in a collaboration with Bristol-Myers Squibb, in an analysis of approximately 1000 patients enrolled in two randomized, double-blind trials (CheckMate 067 and 069) of nivolumab and ipilimumab, used in combination and as monotherapy, for patients with previously-untreated advanced melanoma.

The characterization of TFS was based on TWiST methodology (from the 1990s), which stands for time without symptoms or toxicity. TWiST partitions the patients’ life experience over a fixed period of time, into regions that differ with respect to survival, disease control, treatment administration, and toxicity. The method plots Kaplan-Meier curves for several time-to-event end points, starting with overall survival; and then focuses on the areas under and between the Kaplan-Meier curves. These areas are estimated as the restricted mean times (in our analysis, the time period is restricted to 36 months since randomization.

TFS was defined as the area between the Kaplan-Meier curves for 2 conventional end points, defined from randomization: the time to ICI protocol therapy cessation; and the time to subsequent therapy initiation or death. TFS was then partitioned into 2 further regions, representing TFS spent without and with toxicity from the ICI protocol therapy. TFS is represented by the blue areas in the Figure.

Figure. Estimates of TFS, with and without toxicity, and other health states over the 36-month period since randomization, according to treatment group. Toxicity is defined by grade ≥2 treatment-related AEs.

Over the 36-month period, the restricted-mean TFS was longer for nivolumab plus ipilimumab (11.1 months) compared with nivolumab (4.6 months; difference, 6.5; 95%CI, 5.0-8.0) or ipilimumab (8.7 months; difference, 2.4; 95%CI, 0.8-4.1). The TFS represented 31% (12% with and 19% without toxicity), 13% (4% and 9%, respectively), and 24% (4% and 19%, respectively) of the 36-month period, respectively in the 3 treatment groups. Here, toxicity is defined as persistent or late-onset grade 2 or higher treatment-related AEs. When limited to grade ≥3 treatment-related AEs, TFS with toxicity represented 3%, 1% and <1% of the 36-month period.

With these results in mind, we propose that in addition to the conventional end points of progression-free and overall survival, clinical trials involving immuno-oncology agents should estimate and compare TFS with and without toxicity between different therapeutic strategies in order to more completely capture patient experiences. As development of the TFS model advances, it should facilitate the unified analysis of efficacy, toxicity, quality of life, and cost to ensure that we identify treatments providing the most value for our patients.

There is a lot of discussion about adverse events that occur during and after ICI therapy. A surprising result was to see that persistent grade ≥3 treatment-related AEs made up a small proportion of the TFS period for all treatments.

Next Steps and Application of Findings into Practice

The idea of TWiST and our TFS approach, is to integrate the analysis of treatment administration, toxicity, efficacy, and quality of life, of all patients and facilitate the clinician-patient discussion of the trade-offs of different therapies.

We continue development of the TFS approach, with the aim to facilitate the unified analysis of efficacy, toxicity, and quality of life. Because this analysis was a proof-of-principle and our results of the CheckMate 067 and 069 trials must be interpreted within the framework of trial designs, we intend to estimate the TFS outcome and apply the method in different clinical trial scenarios.

We recently reported on TFS in patients with treatment-naïve, advanced clear cell renal cell carcinoma who participated in the randomized CheckMate 214 trial of nivolumab + ipilimumab compared with sunitinib (again, this work is in collaboration with Bristol-Myers Squibb).6 This trial provided the opportunity to examine TFS in a clinical trial comparison of ICI vs systemic targeted therapy, which has greater contrast in the durations of therapy and adverse events. In this work, we added an area for toxicity on the protocol treatment as well.

Ideally, we aim to compare TFS in clinical trials exploring combination regimens with cessation of therapy after a fixed or maximal duration (eg, 2 years), or cessation after achievement of a clinical milestone such as a complete or near-complete response, or when the trade-offs of differing doses are under investigation.

Also, the TFS concept assumes quality-of-life varies while on and off anticancer therapy and is better during TFS without toxicity. Determining the extent to which quality-of-life returned to baseline during TFS and differed with and without toxicity is also an intended area of investigation. This may also provide greater insight into a functional definition of toxicity for integration into the analysis.

This work represents academic-industry collaboration, and we received a grant from Bristol-Myers Squibb for the project (and ongoing projects). The investigators’ time was also supported by multiple NIH grants.

References

1. Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol. 1999;17(7):2105-2116.

2. Korn EL, Liu P-Y, Lee SJ, et al. Meta-analysis of phase II cooperative group trials in metastatic stage IV melanoma to determine progression-free and overall survival benchmarks for future phase II trials. J Clin Oncol. 2008;26(4):527-534.

3. Wolchok JD, Chiarion-Sileni V, Gonzalez R, et al. Overall survival with combination nivolumab and ipilimumab in advanced melanoma. N Engl J Med. 2017;377:1345-1356.

4. Hodi FS, Chiarion-Sileni V, Gonzalez R, et al. Nivolumab plus ipilimumab or nivolumab alone versus ipilimumab alone in advanced melanoma (CheckMate 067): 4-year outcomes of a multicentre, randomised, phase 3 trial. Lancet Oncol. 2018;19:1480-1492.

5. Callahan MK, Kluger H, Postow MA, et al. Nivolumab plus ipilimumab in patients with advanced melanoma: updated survival, response, and safety data in a phase I dose-escalation study. J Clin Oncol. 2018;36:391-398.

6. Regan MM, Atkins MB, Powles T, et al. Treatment-Free Survival, With and Without Toxicity, as a Novel Outcome Applied to Immuno-Oncology Agents in Advanced Renal Cell Carcinoma (aRCC). Presented at: the 2019 ESMO Congress; September 27-October 1, 2019; Barcelona, Spain. Abstract 3367.

Disclosures

The research reported in the Journal of Clinical Oncology was supported by a grant from Bristol-Myers Squibb; editorial assistance was provided by Kakoli Parai, PhD, and Andrea Lockett at StemScientific, an Ashfield Company, funded by Bristol-Myers Squibb.

Dr Regan

Consulting or Advisory Role: Ipsen (Inst), Merck, Bristol-Myers Squibb

Research Funding: Veridex (Inst), OncoGenex (Inst), Pfizer (Inst), Ipsen (Inst), Novartis (Inst), Merck (Inst), Ferring (Inst), Celgene (Inst), AstraZeneca (Inst), Pierre Fabre (Inst), Bayer AG (Inst), Bristol-Myers Squibb (Inst), Roche (Inst), Astellas Pharma (Inst), Medivation (Inst), Janssen Pharmaceuticals (Inst), Millennium Pharmaceuticals (Inst), Sanofi (Inst), Sotio (Inst), Dendreon (Inst), TerSera (Inst)

Travel, Accommodations, Expenses: Bristol-Myers Squibb

Dr McDermott

Consulting or Advisory Role: Bristol-Myers Squibb, Merck, Genentech, Roche, Pfizer, Exelixis, Novartis, X4 Pharma, Array BioPharma, Peloton Therapeutics, EMD Serono, Jounce Therapeutics, Alkermes, Eli Lilly

Research Funding: Prometheus Laboratories (Inst), Bristol-Myers Squibb (Inst), Merck (Inst), Genentech (Inst), Novartis (Inst), Alkermes (Inst), Peloton Therapeutics (Inst)

Other Relationship: Beth Israel Deaconess Medical Center