Dynamics of clinical biomarkers as predictors of immunotherapy (IT) benefit in metastatic melanoma (MM) patients (pts) treated in reference institu...

Date 24 November 2018
Event ESMO Asia 2018 Congress
Session Poster display - Cocktail
Topics Melanoma
Personalised/Precision Medicine
Immunotherapy
Presenter Alberto Hernando
Citation Annals of Oncology (2018) 29 (suppl_9): ix105-ix108. 10.1093/annonc/mdy439
Authors A. Hernando1, R. Dienstmann2, C. Ortiz Velez1, G. Villacampa2, F.J. Ros Montañá1, J.F. Grau Béjar3, A. Garcia-Alvarez2, P.G. Nuciforo4, A. Gros5, V. Garcia-Patos6, D. Bodet6, B. Ferrer Fabrega6, J. Recio6, E. Muñoz-Couselo3
  • 1Medical Oncology, Vall d'Hebron University Hospital, 08035 - Barcelona/ES
  • 2Oncology Data Science, Vall d'Hebron University Hospital, 08035 - Barcelona/ES
  • 3Medical Oncology Department, Vall d'Hebron University Hospital, 08035 - Barcelona/ES
  • 4Molecular Oncology, Vall d'Hebron University Hospital, 08035 - Barcelona/ES
  • 5Tumor Immunology And Immunotherapy, Vall d`Hebron Institute of Oncology (VHIO)-Cellex Center, 08035 - Barcelona/ES
  • 6Dermatology Department, Vall d'Hebron University Hospital, 08035 - Barcelona/ES

Abstract

Background

Baseline LDH and derived neutrophil–lymphocyte ratio (dNLR) are markers of response to IT. Cutaneous immune-related adverse events (irAEs) are also associated with improved outcomes in MM patients. We hypothesized whether dynamic shifts in LDH, dNLR and incidence of irAEs may impact prognosis of MM pts treated with IT.

Methods

We analyzed 52 MM pts treated at VHIO with single agent IT Anti-PD-1 (Nivolumab or Pembrolizumab) or Anti-CTLA-4 (Ipilimumab) outside clinical trials. DNLR and LDH were collected at baseline and after cycle 2 of IT. IrAEs were assessed with CTCAE v.4.0. Primary endpoint was immunotherapy overall survival (IOS). A cutoff value of dNLR ≥2.5 and LDH ≥1.5xULN was set by the maximization of the log-rank test. Changes in dNLR, LDH and incidence of irAEs were correlated with IOS.

Results

Out of 52 pts, median age was 62.8 years, 49 (96%) ECOG ≤1, treated with Ipilimumab 14 (27%), Nivolumab 20 (38%) or Pembrolizumab 18 (35%). BRAF mutation was present in 19 (37%) pts. Median prior lines were 1 (0-3) and median follow-up was 14.4 months. Baseline dNLR ≥2.5 and LDH ≥1.5xULN were associated with worse IOS (HR: 5.7; 2.2 – 14.3; p < 0.001 and HR: 1.9; 0.8 – 4.2; p = 0.13, respectively). At cycle 2 assessment, dNLR ≥2.5 and LDH ≥1.5xULN also resulted in worse IOS (HR: 29.2; 7.4 – 114.6; p < 0.001 and HR: 3.9; 1.5 – 10.1; p < 0.01 respectively). Importantly, pts switching from either high dNLR to low dNLR (HR: 0.14; 0.03 – 0.74; p = 0.02) or high LDH, to low LDH (HR: 0.08; 0.01 – 0.68; p = 0.02) had significantly better IOS than those with high scores at cycle 2. In addition, significantly longer IOS was also observed for ≥G2 irAEs development (HR: 0.2; 0.05 – 0.89; p = 0.03) in a survival model adjusted for time-dependent covariates.

Conclusions

LDH and dNLR at baseline may impact on prognosis, however in our cohort both cycle 2 measures and shifts from baseline to cycle 2 scores better defined IOS in MM pts. These results suggest that laboratory changes may help indentify “super-responders” or “hyperprogressors” and therefore optimize intrapatient prognostic estimations. We also identified irAEs as a strong predictor of long-term improved outcomes.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

VHIO.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.