br Impact of Adjuvant Chemotherapy
Impact of Adjuvant Chemotherapy on PD-L1 Expression
Figure 1 Changes in Programmed Death-ligand 1 (PD-L1) Expression From Initial Stage to Recurrence in Nonesmall cell Lung Cancer. Four Immunohistochemistry Score Groups Were Defined: TC0 < 1%, TC1 ‡ 1% and < 5%, TC2 ‡ 5% and < 50%, and TC3 ‡ 50%. Upregulation of PD-L1 Expression From Initial Tumor Sample ([TC0) (A) and Biopsy ([TC3) (B) after Adjuvant Therapy. Increase in PD-L1 Expression From Initial Tumor Sample ([TC1) (C) and Biopsy ([TC3) (D) After Adjuvant Therapy. Change in PD-L1 Expression in the Initial Tumor Sample ([TC2) (E) and Biopsy ([TC3) (F) at Recurrence With No Adjuvant Chemotherapy
patients with KRAS mutations towards standard chemotherapy in second-line treatment.28 More recently, Jeanson et al showed that for patients with KRAS-mutant NSCLC, the efficacy of immune checkpoint inhibitors is similar to that for patients with other types of NSCLC, yet PD-L1 AUY922 (NVP-AUY922) seems to be more relevant for predicting the efficacy of immune checkpoint inhibitors in KRAS-mutant NSCLC.29
The major limitations of our study are the small sample size and its retrospective and exploratory nature. Despite consider-able efforts, we were only able to retrieve 72 tissue samples that matched initial tumor samples and recurrences. The main reason was the lack of a biopsy at NSCLC recurrence. Secondly, often only fine needle biopsy was performed at tumor recurrence, which yields limited tumor material unsuited for PD-L1 anal-ysis. In addition, harmonization studies regarding PD-L1 anal-ysis on biopsy and resection specimens are still lacking up to this date. Intratumoral heterogeneity of PD-L1 expression in NSCLC has been verified in recent studies, underlining the fact that small tissue samples such as biopsy specimens may cause false-negative results and may not be representative of the entire tumor specimen.30
However, despite the small sample size of our study, we were able to create 2 homogenous study groups consisting of all patients with the diagnosis of adenocarcinoma who underwent initial surgical
resection followed or not by adjuvant chemotherapy with no pre-vious exposure to radiotherapy. Our study strongly suggests the impact of chemotherapy on PD-L1 expression and the conse-quences thereon.
Table 2 Upregulation in PD-L1 Expression for the Indicated Cases After Adjuvant Chemotherapy and Correlation With Oncogenic Driver Mutations: PD-L1 Expression (TC Score)
EGFR ALK KRAS Pre, % Post, %
Abbreviations: ALK ¼ anaplastic lymphoma kinase; EGFR ¼ epidermal growth factor receptor; PD-L1 ¼ programmed death-ligand 1; WT ¼ wild type.
Max Lacour et al
Figure 2 (A) Changes in PD-L1 Expression From Initial Tumor to Recurrence Without Adjuvant Chemotherapy. (B) Upregulation in PD-L1 Expression From Initial Tumor to Recurrence after Adjuvant Chemotherapy
Abbreviation: PD-L1 ¼ programmed death-ligand 1.
We have demonstrated the variability in PD-L1 expression be-tween primary NSCLC and tumor recurrence and the importance of re-biopsy at tumor relapse.
Chemotherapy might induce PD-L1 expression in NSCLC, which supports the idea that combination therapy of standard chemotherapy with immune checkpoint inhibitors such as PD-1/ PD-L1 inhibitors can have good anti-tumoral effect in patients with NSCLC.
Given the small sample size of our preliminary data, future studies with larger number of patients are necessary to confirm our promising data and eventually improve the clinical outcome of patients with NSCLC.
Clinical Practice Points
Platinum-based chemotherapy might increase PD-L1 expression in NSCLC specimens, supporting the addition of immuno-therapy such as PD-1/PD-L1 inhibitors as a new therapeutic strategy for advanced or metastatic NSCLC.
PD-L1 expression is a dynamic biomarker: even if the initial tumor is PD-L1-negative, the tumor recurrence might be PD-L1-positive, possibly justifying re-biopsy at recurrence.
More patients than expected might benefit from immunotherapy at relapse.
KRAS alterations correlate with PD-L1 expression in lung adenocarcinoma; therefore, blockade of the PD-1/PD-L1 pathway by immune checkpoint inhibitors might be a prom-ising therapeutic option for KRAS-mutated NSCLC.
The authors thank Susanne Dettwiler from the Institute of Surgical Pathology for helping with the staining of the slides, as well as the “Stiftung für angewandte Krebsforschung” for their support.