91PD_PR - Response to salvage chemotherapy following exposure to PD-1/PD-L1 inhibitors in patients with NSCLC

Date 07 May 2017
Event ELCC 2017
Session Targeted therapies and immunotherapies
Topics Immunotherapy
Thoracic Malignancies
Non-Small Cell Lung Cancer
Presenter Sacha Rothschild
Citation Annals of Oncology (2017) 28 (suppl_2): ii28-ii51. 10.1093/annonc/mdx091
Authors S.I. Rothschild1, P. Leger2, E.L. Castellanos2, R.N. Pillai3, S.J. York2, L. Horn2
  • 1Department Internal Medicine, Medical Oncology, Universitätsspital Basel, 4031 - Basel/CH
  • 2Vanderbilt University, Nashville/US
  • 3Winship Cancer Institute, Emory University, Atlanta/US



Immune checkpoint inhibitors are active for patients with stage IV NSCLC who have progressed following platinum-based chemotherapy. We evaluated responses to chemotherapy in patients with NSCLC who had progressed on a checkpoint inhibitor.


Eligible patients were adults with NSCLC who received salvage chemotherapy following PD-1/PD-L1 inhibitors (cases) versus no PD-1/PD-L1 inhibitors (controls). CT-imaging was done within 4 weeks of initiation of salvage chemotherapy and every 6 weeks thereafter. Revised RECIST guidelines were used to define response. Clinical and imaging data were abstracted from review of electronic medical records. Multivariate logistic regression analysis was used to calculate probability of response.


355 patients’ charts were reviewed and 82 patients met eligibility criteria. Among evaluable patients, 46 were males versus 36 females. 67 patients were classified as cases versus 15 controls. 56 patients received nivolumab, 7 pembrolizumab and 4 atezolizumab. 63 (77%) patients had adenocarcinoma, 18 (22%) squamous cell carcinoma and 1 (1%) large cell carcinoma. The mean number of chemotherapy regimens prior to salvage chemotherapy was 2.37 (95% CI: 2.10-2.64) in cases versus 1.93 (95% CI: 1.32-2.54) in controls. Salvage drugs used included docetaxel (62%), pemetrexed (20%), gemcitabine (12%) and paclitaxel (6%). 18 (27%) cases had partial response to chemotherapy versus 1 (7%) controls. 15 (22%) cases had progressive disease versus 6 (40%) controls. 34 (51%) cases had stable disease versus 8 (53%) controls. The odds ratio for achieving a partial response was 0.30 (95% CI: 0.18 to 0.50, p = 0.000). In multiple logistic regression model, age, gender, number of prior chemotherapy regimens, tumor histology, smoking status, different salvage chemotherapy regimens were not associated with the likelihood of achieving a partial response.


The odds of achieving a partial response to salvage chemotherapy were significantly higher in patients with prior exposure to PD-1/PD-L1 inhibitors. This observed difference however warrants confirmation in larger cohorts. Ongoing investigations include the duration of response as well as evaluation of toxicity.

Clinical trial identification

Legal entity responsible for the study





All authors have declared no conflicts of interest.