Dr. Weiss is Assistant Professor (Medical Oncology), Developmental Therapeutics, Melanoma Program, Yale Cancer Center. Dr. Sznol is Professor of Medicine (Medical Oncology); Co-Director, Cancer Immunology Program, Yale Cancer Center Co-Director, Yale SPORE in Skin Cancer, New Haven, Connecticut.
Sarah A. Weiss, MD
Mario Sznol, MD
DECISIONS TO CONTINUE treatment of anticancer agents in patients with metastatic disease are traditionally based on intermittent response assessments. A commonly held assumption for many years, particularly prior to the introduction of targeted and immune therapies, was that evidence of disease progression using standard response criteria indicated loss of clinical activity and therefore loss of benefit for the therapeutic agent, thus requiring a treatment change. Prior response criteria, including the Response Evaluation Criteria in Solid Tumors (RECIST), defined disease progression as a threshold increase in measurements of preselected lesions, unequivocal disease progression in nontarget lesions, or the development of new lesions.1
Although the question has not been formally tested, clinical experience suggests that at least for certain targeted agents, treatment beyond disease progression may provide additional clinical benefit, because disease progression on targeted therapy may still be slower than on an alternative therapy. The same may be true, but is unknown, for certain types of disease progression with cytotoxic drugs.
Atypical Response Patterns With Immunotherapy
IMMUNOTHERAPY AGENTS, in particular the immune checkpoint inhibitors such as anticytotoxic T-lymphocyte– associated protein 4 (anti–CTLA-4) or anti–programmed cell death protein 1 (anti–PD-1) agents, present a particular challenge for oncologists because of the atypical patterns of response seen in a subset of patients. For example, pseudoprogression, defined as initial radiographic tumor growth followed by regression, occurs in 10% to 15% of patients with metastatic melanoma treated with PD-1/programmed cell death ligand 1 (PD-L1)–pathway inhibitors. These atypical patterns of response, caused by either inflammation from tumor immune cell infiltration or a delayed antitumor immune response, could lead to premature discontinuation of an effective therapy for an individual and potential reduction in optimal benefit.2-5
Additionally, if RECIST is used to assess activity in a clinical trial of an agent demonstrating atypical patterns of response that occur in a patient enrolled on a clinical trial, the overall activity of the experimental regimen may be underestimated. After recognition of the atypical response patterns, treatment beyond RECIST-defined disease progression was permitted in many immunotherapy clinical trials and is commonly employed in the standard-of-care administration of these agents.
Pooled Analysis of Treatment Beyond Disease Progression
TO BETTER DEFINE the value of treatment beyond disease progression, Beaver et al reported a retrospective pooled analysis of data from 2,624 patients with metastatic melanoma treated with PD-1 inhibitors.6 Half of patients with RECIST-defined disease progression (692 of 1,361) were treated beyond disease progression. Initial RECIST-defined disease progression most commonly occurred at the first assessment, with the majority at or before the 12-week time point. Patients selected for treatment beyond disease progression tended to have either radiographic disease progression of target lesion(s) or the appearance of a new lesion as the main reason for disease progression.
“Atypical patterns of response [to immunotherapy]…could lead to premature discontinuation of an effective therapy for an individual and potential reduction in optimal benefit.”— Sarah A. Weiss, MD, and Mario Sznol, MD
Tweet this quote
In contrast, patients not selected for treatment beyond disease progression had a combination of factors responsible for initial progressive disease, including the growth of target and nontarget lesions and the appearance of new lesions. At baseline, demographic characteristics for patients ultimately selected for treatment beyond disease progression were not meaningfully different from those of patients who were not treated beyond disease progression. However, at the time of RECIST disease progression, more patients treated beyond disease progression had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 (67% vs 34%) and normal lactate dehydrogenase (LDH) levels (63% vs 31%).
The characteristics of patients in the treatment beyond disease progression group unsurprisingly reflect common selection criteria used to guide treatment beyond disease progression decisions in current clinical practice. Patients whose disease progresses on PD-1 inhibitors are considered for treatment beyond disease progression if they maintain a good performance status, are deemed by the treating physician as receiving benefit from treatment, and are tolerating therapy (especially in the absence of other effective therapeutic options). Conversely, rapid clinical decline, evidence of extensive or rapid tumor growth, and/or toxicity would be relative contraindications. The rates of serious and immune-related adverse events were not higher in the treatment beyond disease progression group than in the no treatment beyond disease progression group in this analysis, and therefore ongoing treatment with PD-1 inhibitors did not jeopardize patient safety.
Responses after treatment beyond disease progression were judged in relation to the tumor burden at baseline before anti–PD-1 therapy and at the time of initial disease progression. Eleven percent of patients (57 of 531) treated beyond disease progression achieved at least a 30% reduction in tumor burden in relation to baseline, which largely occurred when disease progression was due to the appearance of a new lesion alone. The analyses did not address what proportion of patients underwent surgical resection or other local therapy for a single site of growth or a single new lesion and how this impacted the response data. This is particularly important because 45% of the patients who responded to treatment beyond disease progression had the development of a new lesion(s) as the sole reason for initial disease progression.
Overall, 19% of evaluable patients (95 of 500) treated beyond disease progression achieved at least a 30% reduction in tumor burden relative to their tumor burden at the time of RECIST-defined disease progression. Of the 95 patients, 17% developed their initial RECIST disease progression more than 24 weeks after starting therapy. For those patients treated beyond disease progression, the median response time was relatively quick, at 2.8 months, and the median duration of treatment was 9.4 months.
Shortcomings of RECIST-Defined Disease Progression
THIS RETROSPECTIVE analysis highlights several additional shortcomings of disease progression as defined by RECIST. Notably, 284 of 692 patients (41%) in the treatment beyond disease progression group already had a reduction in their target lesion tumor burden from baseline despite being labeled as having RECIST-defined disease progression, often due to the appearance of a new lesion or progression of a target lesion, compared with 174 of 669 patients (25%) in the no treatment beyond disease progression group. Additionally, post–disease progression responses were also observed in patients who did not receive treatment beyond disease progression. Although 64 of 669 patients in the no treatment beyond disease progression group had subsequent tumor assessments available for analysis, 10 of the 64 (16%) demonstrated post–disease progression responses before alternative therapies were started. It is unknown whether continued PD-1 inhibitor treatment could have benefited this group further, potentially delaying or obviating the need for transition to another therapy.
There was superior median overall survival in the treatment vs no treatment beyond disease progression groups (24.4 vs 11.2 months), but these results are possibly explained by differences in prognostic features or the unknown underlying biologic differences between the groups. In CheckMate 066, the sole trial analyzed in which treatment beyond disease progression was allowed in both the treatment (nivolumab [Opdivo]) and control (dacarbazine) groups, there were no differences in 1-year overall survival rates for patients treated or not treated beyond disease progression in the nivolumab arm; patients in the dacarbazine arm survived longer if chemotherapy was continued beyond disease progression, compared with those who did not continue dacarbazine beyond disease progression.
Search for Clear Answer Continues
ULTIMATELY, THE very elegant analyses performed by Beaver et al do not provide a clear answer on the value of treatment beyond RECIST disease progression for patients with metastatic melanoma receiving anti–PD-1 or anti–PD-1 plus anti–CTLA-4 agents. Substantial tumor regression can occur after RECIST disease progression in some patients, even without additional therapy; the analyses did not address how often the additional regression occurs in lesions stable or regressing at the time of RECIST disease progression, vs regression in growing areas or new lesions.
“New response criteria specific to the mechanisms of immunotherapy should be implemented in clinical trials and compared with RECIST to refine how disease progression is defined, because major treatment decisions largely hinge on this label of disease progression.”— Sarah A. Weiss, MD, and Mario Sznol, MD
Tweet this quote
It should also not be assumed that a 30% regression after RECIST disease progression is the optimal surrogate for additional benefit of continued treatment; perhaps smaller amounts of tumor regression or even stable disease beyond RECIST disease progression may translate into improved survival for an individual patient. Because continued treatment does not markedly increase the risk for toxicity, a perceived potential benefit may be the predominant deciding factor in continuing treatment beyond disease progression and may impede the conduct of definitive randomized trials to answer the question.
The results may also vary by disease type. For example, in a U.S. Food and Drug Administration analysis of 535 patients with metastatic non–small cell lung cancer treated on clinical trials with anti– PD-1 therapy in which treatment beyond disease progression was allowed, the regression rate was just 8% (10 of 121) in the subset of patients who received treatment beyond disease progression.7 The debate could also be informed by randomized trials to determine the value of continuing treatment in patients with stable disease or objective response after a defined treatment period, for example 6 months or 1 year.
Only a small subset of patients appears to develop tumor regression after RECIST disease progression. The data presented by Beaver et al suggest that patients with an ECOG performance status of 0, a normal LDH level, evidence of tumor regression in the context of overall RECIST disease progression, and disease progression limited to a new lesion or a target or nontarget lesion, are most likely to demonstrate tumor regression of at least 30% at subsequent scans with continued treatment. Remarkably, this subsequent tumor regression was observed in patients who had RECIST disease progression beyond 24 weeks after starting treatment. The latter observation has important implications for interpreting the results of trials adding an investigational agent to anti–PD-1 treatment at the time of RECIST disease progression.
“The majority of patients treated beyond disease progression will not have subsequent responses and will require new treatment strategies to overcome resistance to PD-1 inhibitors.”— Sarah A. Weiss, MD, and Mario Sznol, MD
Tweet this quote
Adaptations of Radiographic Response Criteria
NUMEROUS ADAPTATIONS of radiographic response criteria have been developed to address the concern that RECIST-defined responses underestimate the survival benefit from PD-1 inhibitors. For example, immune-related response criteria measure new lesions as part of the total tumor burden and require radiographic confirmation of disease progression.3,5,8 Immunotherapy RECIST allow unconfirmed disease progression before a complete response, partial response, or stable disease is deemed to have occurred9; and most recently immune-modified RECIST do not include the growth of nontarget lesions in the definition of disease progression and also allow for best overall responses, even after radiographic progressive disease has occurred.10 New response criteria specific to the mechanisms of immunotherapy should be implemented in clinical trials and compared with RECIST to refine how disease progression is defined, because major treatment decisions largely hinge on this label of disease progression.
The majority of patients treated beyond disease progression will not have subsequent responses and will require new treatment strategies to overcome resistance to PD-1 inhibitors, and this should be explained to patients who consent to treatment beyond disease progression. The refinement of radiographic response criteria will be an important step, but eventually identification and development of predictive tumor and/or serum markers must also be incorporated to suggest which patients will most benefit from treatment with PD-1 inhibitors beyond disease progression. ■
DISCLOSURE: Dr. Sznol has stock and other ownership interests in Amphivena, Intensity Therapeutics, and Actym Therapeutics; is a consultant or advisor to Hinge, Allakos, AbbVie, Torque, Genmab, Symphony Evolution, Molecular Partners, NewLink Genetics, Inovio Pharmaceuticals, Seattle Genetics, Omniox, Modulate, Theravance, Lycera, Adaptimmune, Biodesix, Merck Sharp & Dohme, Lilly, Novartis, Nektar, Kyowa Hakko Kirin, AstraZeneca/MedImmune, Genentech/Roche, BMS, Baxalta/Shire, and Pierre Fabre; and has other relationships with Haymarket Media, Research to Practice, TRM Oncology, Physician Education Resource, Prime Oncology, Vindico, Clinical Care Options, DAVA Oncology, Academic CME, and Imedex. Dr. Weiss reported no conflicts of interest.
REFERENCES
1. Eisenhauer EA, Therasse P, Bogaerts J, et al: New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1). Eur J Cancer 45:228- 247, 2009.
2. Long GV, Weber JS, Larkin J, et al: Nivolumab for patients with advanced melanoma treated beyond progression: Analysis of 2 phase 3 clinical trials. JAMA Oncol 3:1511-1519, 2017.
3. Wolchok JD, Hoos A, O’Day S, et al: Guidelines for the evaluation of immune therapy activity in solid tumors: Immune-related response criteria. Clin Cancer Res 15:7412-7420, 2009.
4. Chiou VL, Burotto M: Pseudoprogression and immune-related response in solid tumors. J Clin Oncol 33:3541-3543, 2015.
5. Hodi FS, Hwu WJ, Kefford R, et al: Evaluation of immune-related response criteria and RECIST v1.1 in patients with advanced melanoma treated with pembrolizumab. J Clin Oncol 34:1510-1517, 2016.
6. Beaver JA, Hazarika M, Mulkey F, et al: Patients with melanoma treated with an anti-PD-1 antibody beyond RECIST progression: A US Food and Drug Administration pooled analysis. Lancet Oncol 19:229-239, 2018.
7. Kazandjian D, Keegan P, Suzman DL, et al: Characterization of outcomes in patients with metastatic non-small cell lung cancer treated with programmed cell death protein 1 inhibitors past RECIST version 1.1-defined disease progression in clinical trials. Semin Oncol 44:3-7, 2017.
8. Nishino M, Giobbie-Hurder A, Gargano M, et al: Developing a common language for tumor response to immunotherapy: Immune-related response criteria using unidimensional measurements. Clin Cancer Res 19:3936-3943, 2013.
9. Seymour L, Bogaerts J, Perrone A, et al: iRECIST: Guidelines for response criteria for use in trials testing immunotherapeutics. Lancet Oncol 18:e143-e152, 2017.
10. Hodi FS, Ballinger M, Lyons B, et al: Immune-modified Response Evaluation Criteria In Solid Tumors (imRECIST): Refining guidelines to assess the clinical benefit of cancer immunotherapy. J Clin Oncol 36:850-858, 2018.
