Search

e15062 Background: Conventional anticancer drugs have been developed to improve overall survival (OS) and/or progression-free survival (PFS) in the palliative intent setting. However, long-term survival is often limited by acquired biologic resistance. In contrast to conventional anticancer drugs, ICIs putatively preserve long-term survival in a small subset of patients. However, in the absence of a gold-standard definition, long-term survival improvement has not been adequately quantified. Long-term survival rate difference has been proposed as a method of quantifying long-term survival. However, there a paucity of literature demonstrating statistical improvements in survival rate. Therefore, the objective of the study was to quantify the improvement in long-term survival rate in OS and PFS improvement in ICIs compared with non-ICIs. Methods: The FDA Hematology/Oncology Approvals & Safety Notification web page was searched for RCTs cited in oncology drug approvals between 01-2011 and 06-2019. The most recent OS and PFS updates were used when available. OS and PFS curves were digitized to reconstruct pseudo individual patient data as per established methods. Survival rate improvement at 1, 2 and 3 years were calculated as differences between the experimental arm and control arm. Survival rate improvements across RCTs at 1, 2 and 3 years were pooled using a random-effects model and the results were compared between ICI and non-ICIs. Subgroup analyses were conducted for melanoma and non-small cell lung cancer. Results: We identified 130 RCTs (22 ICIs, 108 non-ICIs) for analysis. The absolute OS rate improvement was 9.5% in ICIs and 4.9% in non-ICIs at 1-year (difference = 4.6%, p < 0.001), 10.3% in ICIs and 4.9% in non-ICIs at 2-years (difference = 5.3%, p < 0.001), and 10.2% in ICIs and 5.1% in non-ICIs at 3-years (difference = 4.6% p 0.006). The absolute PFS rate improvement was 10.3% in ICIs and 17.7% in non-ICIs at 1-year (difference = -7.3% p < 0.001), and 8.9% in ICIs and 15.5% in non-ICIs at 2-years (difference = -6.6% p < 0.001). Conclusions: Long-term OS survival rate improvements, while statistically significant were only modestly higher in ICIs compared to non-ICIs. PFS survival rate differences were statistically lower in ICIs compared to non-ICIs. While ICIs represent an important shift towards long-term survival, greater emphasis should be placed on rigorously evaluating long-term survival rate improvement and reporting of future novel therapies.

2585 Background: Recently, anti-cancer agents have generated excitement due to their capacity to preserve long-term survival in some patients, represented by a “tail of the survival curve”. However, as traditional measures of clinical benefit may not accurately capture long-term survival, amendments to various valuation frameworks have been proposed to capture this benefit. The purpose of this study was to determine how frequently immune checkpoint inhibitor vs. non-immune checkpoint inhibitor anti-cancer agents, displayed trends of long-term survival, as defined by the American Society of Clinical Oncology Value Framework (ASCO-VF) and European Society of Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS), as well as to analyze the degree of agreement between ASCO and ESMO frameworks. Methods: Anti-cancer agents from phase II or III randomized controlled trials (RCTs) cited for clinical efficacy evidence in drug approval by the Food and Drug Administration (FDA) between January 2011 and March 2018 were identified. Data required for ASCO-VF and ESMO-MCBS were extracted. Difference in how often long-term survival bonuses were awarded were calculated in all RCTs, as well as immune checkpoint inhibitor and non-immune checkpoint inhibitor RCTs individually. Cohen’s Kappa statistic was calculated to evaluate agreement between ASCO-VF and ESMO-MCBS. Results: 100 RCTs were analyzed. RCTs were awarded ASCO-VF version 2 (v2) “tail of the curve” bonuses more often than ESMO-MCBS version 1.1 (v1.1) “immunotherapy-triggered” long-term plateau adjustments (45% vs. 2.6%). Comparing to non-immune checkpoint inhibitor RCTs, immune checkpoint inhibitor RCTs were not more likely to receive ASCO-VF v2 bonuses/ESMO-MCBS v1.1 adjustments (p = 0.32/ p = 0.40). Long-term survival agreement between the two frameworks was poor (kappa: 0.01; p = 0.50). Conclusions: The ASCO-VF v2 and ESMO-MCBS v1.1 may require additional refinement in order to accurately capture the benefit of long-term survival or immune checkpoint inhibitor and non-immune checkpoint inhibitor agents may not preserve substantially different long-term survival populations.