Further analyses of biomarkers

Immune checkpoint inhibitors have become established as standard therapy for certain subsets of NSCLC patients. Efforts are ongoing to optimise the benefits gained through these drugs, by identification of reliable prognostic and predictive biomarkers, such as PD-L1 expression in tumour cells or infiltrating immune cells, CD8-positive tumour-infiltrating lymphocytes, smoking status, and mutation burden. The necessity for collaboration in this field is obvious; as yet, progress in the area of immuno-oncological biomarkers has been impeded by marked variations with regard to development and reporting (e.g., different cut-offs).

Targets of immune checkpoint inhibitors

An analysis of the tumour samples of 208 patients with NSCLC assessed the prognostic relevance of PD-L1, PD-1 and CTLA-4 [1]. PD-L1 protein expression in tissue microarrays of archival tumour samples was tested using IHC. Mutations, expression, and copy-number variations in the PD1, PD-L1 and CTLA-4 genes were retrieved from the publicly available The Cancer Genome Atlas (TCGA) database. The scientists estimated the differences in disease-free survival and OS between patients with high and low PD-L1 protein expression, using univariate and multivariate survival analyses. The prognostic impact of genetic alterations was assessed in the overall population and in subsets of adenocarcinoma, squamous cell carcinoma, smokers and non-smokers.

The analysis showed that there is higher PD-L1 expression in smokers, non-squamous carcinoma, and females, with no significant differences found according to race and tumour stage or grade. Higher normalised PD-1 gene expression correlated with improved patient survival. Mutations of the PD-1, PD-L1 and CTLA4 genes were very rare in NSCLC and showed no association with survival. According to the conclusions of the authors, the PD-L1 protein, but not PD-L1 gene expression, might have value as a prognostic marker in early-stage NSCLC.

PD-L1 and immune infiltrates

There has been interest in combining PD-1/PD-L1 inhibitors with EGFR or ALK TKIs in NSCLC, because data has suggested that there are links between the underlying molecular principles. For instance, a PD-1 block can improve survival in EGFR-mutant mouse models [2]. Several trials involving EGFR TKIs or ALK TKIs and PD-1/PD-L1 inhibitors are currently ongoing.

Therefore, a retrospective analysis investigated PD-L1 expression patterns and immune infiltrates in NSCLC patients with EGFR mutations (n = 68) and ALK rearrangements (n = 26) [3]. This showed that EGFR-mutant and ALK-positive lung cancers can express PD-L1 and have CD8+ immune infiltrates, although most of these tumours do not have both. As the authors noted, this might underlie the low response rates that were seen with PD-1 pathway inhibition in never smokers and light smokers in the CheckMate 057 trial [4]. In a subset of patients, there were changes in PD-L1 expression and immune infiltrates over time and/or following treatment. Future studies with early, ‘on-treatment’ biopsies would be necessary to determine whether immune cells can be recruited to the tumour environment following the initial treatment with TKIs.

Reasons for screening failure

As trials for NSCLC increasingly involve small patient populations expressing specific molecular abnormalities, many studies now require evaluation of a fresh biopsy prior to enrolment. This is not feasible for all research sites. Moreover, the implications of potentially increased screening duration on the interpretation of clinical outcomes have not been completely explored yet.

To identify factors associated with lack of enrolment, researchers reviewed the charts of 268 patients with NSCLC who had consented to participate in a total of 26 trials requiring biopsies for evaluation of certain biomarkers as part of the patient eligibility. Out of these 268 patients, 141 failed this screening and were thus included in the analysis. 24 % of these were eligible based on biomarker status. The results demonstrated that apart from the presence or absence of a specific biomarker, many factors can lead to screening failure. The reasons included worsening performance status in 52.9 %, pursuit of alternate therapy in 8.8%, out-of-range screening laboratory values in 8.8 %, withdrawal of consent in 5.8 %, and other reasons in 23.5% (Figure).

Figure: Reasons for screening failure in patients who had consented to trials involving bioptic biomarker sampling and were eligible according to biomarker status

Figure: Reasons for screening failure in patients who had consented to trials involving bioptic biomarker sampling and were eligible according to biomarker status

Due to disease progression, many patients had worsening performance status at the time of initiation of the study, as well as abnormal laboratory values. The time that elapsed between the signing of the informed consent and the beginning of the trial was approximately 35 days. Whether such a delay in enrolment based on the duration of the screening either excludes patients who would have been more likely to experience early decline or enrols patients who have become more ill during screening needs to be fully explored.


  1. Owonikoko TK et al., PD-L1, PD-1, and CTLA-4 as prognostic biomarkers in resected non-small cell lung cancer. J Clin Oncol 33, 2015 (suppl; abstr 7551)
  2. Akbay EA et al., Activation of the PD-1 pathway contributes to immune escape in EGFR-driven lung tumors. Cancer Discov 2013; 3(12): 1355-63
  3. Gainor JF et al., Clinical correlation and frequency of programmed death ligand-1 (PD-L1) expression in EGFR-mutant and ALK-rearranged non-small cell lung cancer (NSCLC). J Clin Oncol 33, 2015 (suppl; abstr 8012)
  4. Paz-Ares L et al., Phase III, randomized trial (CheckMate 057) of nivolumab (NIVO) versus docetaxel (DOC) in advanced non-squamous cell (non-SQ) non-small cell lung cancer (NSCLC). J Clin Oncol 33, 2015 (suppl; abstr LBA109)
  5. Nameth DJ et al., Evaluating causes of screen failure (SF) in non-small cell lung cancer (NSCLC) clinical trials requiring specific biomarker (BioM) results for enrollment. J Clin Oncol 33, 2015 (suppl; abstr 7517)

For more articles on cancer, lung cancer and respiratory medicine see
and www.inoncology.com