397P - Immune related adverse events (irAEs) in early phase immunotherapy (IO) trials: Implications for recommended phase 2 dose (RP2D) determination

Date 11 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Clinical research
Complications/Toxicities of treatment
Basic Scientific Principles
Presenter Yada Kanjanapan
Citation Annals of Oncology (2017) 28 (suppl_5): v122-v141. 10.1093/annonc/mdx367
Authors Y. Kanjanapan1, D. Day1, M. Butler2, L. Wang3, A.M. Joshua1, D. Hogg2, N.B. Leighl1, A.R.A. Razak1, A.R. Hansen2, S. Boujos4, M.A. Chappell1, K. Chow4, M. Paolo4, B. Sherwin1, L. Stayner4, L. Soultani4, A. Zambrana1, L.L. Siu5, P.L. Bedard2, A. Spreafico2
  • 1Division Of Medical Oncology And Hematology, Princess Margaret Cancer Centre, M5G 2M9 - Toronto/CA
  • 2Division Of Medical Oncology And Hematology, Princess Margaret Cancer Centre, M5G 1Z5 - Toronto/CA
  • 3Biostatistics Department, Princess Margaret Cancer Centre, M5G 2M9 - Toronto/CA
  • 4Drug Development Program, Princess Margaret Cancer Centre, M5G 1Z5 - Toronto/CA
  • 5Division Of Medical Oncology And Hematology, Princess Margaret Cancer Centre, Toronto/CA



IO agents have a unique profile of irAEs. Due to the potential of delayed severe irAEs, we examined if conventional dose-limiting toxicity (DLT) periods may underestimate the rate of clinically significant irAE (csAE), defined as an irAE that required systemic therapy, drug delay or discontinuation.


A retrospective chart review of patients (pts) on early phase IO trials at Princess Margaret Cancer Centre examined severity (CTCAE v4.0), management, timing of onset and resolution of all grade (G) irAEs. A generalized estimating equation model assessed the association between time on treatment (Rx) and csAE, adjusted for duration of IO. Potential predictors of csAEs were assessed.


From 8/2012-9/2016, 239 pts across 21 trials received ≥1 dose of IO (72% single agent; 28% IO-based combination). The most common tumors were melanoma (23%) and lung (18%) cancer. Among 890 total Rx-related irAEs, 93 (10%) were csAEs, including 22 (24%) endocrine, 15 (16%) gastrointestinal [GI], 11 (12%) respiratory, 10 (11%) skin and 9 (10%) hepatic csAEs. Median onset was ≤90 days for hepatic, GI and general (eg fatigue) csAEs, and >90 days for endocrine, respiratory, skin and musculoskeletal csAEs (P = 0.03). One pt with G3 hepatitis and G4 hypophosphatemia met protocol-defined DLT criteria and 27 irAEs fulfilled DLT criteria but occurred after the DLT period. 61 pts had csAEs, with 21 (34%) having >1 csAE. The onset of first csAE was 0-6 weeks (wks) in 30 (49%) pts; wks 7-12 in 16 (26%); wks 13-48 in 10 (16%); and ≥49 wks in 5 (8%) pts. The odds ratio (OR) for the first csAE occurrence within first 6 wks vs ≥ 6 wks was 3.3 (95% CI 2.0-5.6, P = 0.002), accounting for varying Rx duration. After adjustment for time on Rx, csAE correlated only with response on univariate analysis (OR 4.3, 95% CI 2.1-8.9, P 


Risk of first-onset csAE was higher during the initial 6 wks of IO, supporting use of conventional DLT period for dose escalation decision. However, as late csAEs were also seen, RP2D determination should consider the entire temporal course of csAE. Occurrence of csAEs positively correlated with response to IO, relative to time on Rx.

Clinical trial identification

Legal entity responsible for the study

Princess Margaret Cancer Centre




M. Butler: Advisory board for: Merck, Bristol-Myers Squibb, Novartis, Turnstone, EMD Serono, Immunocore. D. Hogg: Advisory board for Roche, Bristol-Myers Squibb, Novartis, EMD Sereno, Merck. N.B. Leighl: Research funding from Novartis. L.L. Siu: Research funding from Bristol-Myers-Squibb, Merck, Novartis, AstraZeneca/Medimmune and Roche-Genentech. P.L. Bedard: Research funding from Novartis, Roche-Genentech. Bristol-Myers-Squibb, AstraZeneca/Medimmune. All other authors have declared no conflicts of interest.