Back Back
Immunotherapy in Lung Adenocarcinoma
My Thoughts on Immunotherapy and PD-L1 Testing in Metastatic Lung Adenocarcinoma

Released: July 08, 2016

Expiration: July 07, 2017

Activity Information

Activity

Progress
1
Course Completed
Continue

Case Presentation
A 46-year-old male former smoker with a 50 pack-year smoking history was recently referred to my clinic to discuss options for second-line therapy following progression on first-line carboplatin and pemetrexed for stage IV lung adenocarcinoma. Reflex molecular testing obtained by his original oncologist revealed a KRAS G12C mutation and negative PD-L1 staining using the FDA-approved companion assay for pembrolizumab (anti–PD-L1 monoclonal antibody 22C3).

There are currently 4 FDA-approved therapies for the second-line treatment of metastatic non-small-cell lung cancer (NSCLC): nivolumab, pembrolizumab, docetaxel (± ramucirumab), and erlotinib. Because KRAS mutations are associated with primary resistance to EGFR inhibitors, I would not recommend erlotinib for this patient. Being PD-L1 negative, he also is not eligible for pembrolizumab, which is currently FDA approved only in patients with tumors showing detectable PD-L1 expression in ≥ 50% of the tumor cells. In the randomized phase III CheckMate 057 trial, the remaining 2 treatment options, nivolumab and docetaxel, were shown to exhibit similar efficacy in patients with nonsquamous NSCLC and PD-L1–negative tumors (< 1% PD-L1 as assessed with the complementary diagnostic test for nivolumab using anti–PD-L1 monoclonal antibody 28-8). However, 54% of patients in the docetaxel arm reported grade 3/4 treatment-related adverse events vs only 10% of patients in the nivolumab arm, which is consistent with the significantly higher rates of potentially serious toxicities, such as neutropenia, febrile neutropenia, and fatigue, known to be associated with docetaxel. Thus, for second-line therapy in this patient, I recommended nivolumab, which provides a similar OS benefit as docetaxel, while exposing the patient to a lower risk for toxicities and perhaps a better quality of life.

A second recent referral to my clinic addresses a question that I am frequently asked: how to optimally treat patients with lung adenocarcinoma whose tumors express PD-L1 but at levels that fall below the FDA-approved ≥ 50% PD-L1 expression cutoff point for pembrolizumab. This patient was a former light smoker with a 12 pack-year history who had progressed after first-line chemotherapy with carboplatin and pemetrexed followed by maintenance pemetrexed. His tumor tested negative for mutations but tested 18% positive for PD-L1, which is below the current cutoff point for pembrolizumab. However, in the KEYNOTE-010 trial, pembrolizumab achieved an OS benefit both in patients with ≥ 50% and ≥ 1% PD-L1 expression vs docetaxel, suggesting that pembrolizumab would be a more effective agent in this patient than chemotherapy. In the absence of approval of pembrolizumab in this setting, I again recommended nivolumab for this patient. However, based on the results of KEYNOTE-010, the lower 1% cutoff point for pembrolizumab will be considered by the FDA, approval of which will expand treatment options for patients with advanced NSCLC.

Figure. KEYNOTE-010: median OS in patients with ≥ 50% and ≥ 1% PD-L1–expressing tumors.

Atezolizumab, which is FDA approved for bladder cancer and under priority review for lung cancer, is another drug that shows promise in the second- or third-line treatment of advanced NSCLC. In the randomized, all-comer phase II POPLAR trial, atezolizumab significantly improved OS vs docetaxel in a PD-L1 expression–dependent manner (PD-L1 ≥ 1% to < 5%, ≥ 5% to 50%, and ≥ 50%, as assessed with yet another IHC assay using anti–PD-L1 monoclonal antibody SP142 assessing both tumor cells and tumor infiltrating cells with different cut-off points). However, similar to nivolumab in CheckMate 057, atezolizumab and docetaxel showed similar efficacy in patients with PD-L1–negative tumors (< 1% PD-L1). The phase III OAK trial evaluating second-line atezolizumab vs docetaxel in unselected patients stratified by PD-L1 expression has concluded enrollment with results expected later this year. The results of these studies will hopefully lead to the FDA approval of atezolizumab for advanced NSCLC, whether in the setting of minimal PD-L1 expression or regardless of PD-L1 expression is currently unclear.

At this time, I do not routinely assess PD-L1 in my own practice as the PD-1 and PD-L1 inhibitors have demonstrated clinical benefit in patients with both PD-L1–positive and PD-L1–negative tumors, with a response rate of approximately 30% and 10%, respectively. I do consider testing for PD-L1 expression in patients who progress on previous therapy if they are never-smokers or are positive for EGFR mutations or ALK rearrangements and are not a candidate for additional targeted therapy. This is because of the subgroup analyses of CheckMate 057 and KEYNOTE-010 showing these patients have inferior responses to immune checkpoint inhibitors compared to chemotherapy. If their tumors are negative for PD-L1 I would consider chemotherapy but if ≥ 1% of tumor cells stain positive for PD-L1 they may benefit from pembrolizumab or nivolumab. We are also working on optimizing testing for PD-L1 expression at the time of diagnosis in anticipation of the results from the phase III KEYNOTE-024 (patients with advanced NSCLC who have ≥ 50% PD-L1 expression) and CheckMate 026 trials evaluating first-line pembrolizumab and nivolumab, respectively, vs platinum-based chemotherapy in PD-L1–positive advanced NSCLC. A recent press release indicated that KEYNOTE-024 met its primary endpoint of improvement in PFS and OS and I am looking forward to seeing the data later this year.

PD-L1 Assay Standardization
Currently, distinct PD-L1 IHC assays are being used for each individual anti–PD-1 (nivolumab, pembrolizumab) or anti–PD-L1 (atezolizumab, durvalumab) agent as indicated above. The FDA-approved companion assay for pembrolizumab and the complementary diagnostic test for nivolumab use anti–PD-L1 monoclonal antibodies 22C3 and 28-8, respectively. The assays for the investigational agents atezolizumab and durvalumab use anti–PD-L1 monoclonal antibodies SP142 and SP263, respectively. Efforts to standardize PD-L1 testing led by Dr. Fred Hirsch through the IASLC Blueprint Project showed very similar staining for the nivolumab, pembrolizumab, and durvalumab assays, with some small differences for the atezolizumab assay. Thus, it appears that the read‑out for PD-L1–positive vs PD-L1–negative tumors is very similar for the majority of the assays. Finally, tumor heterogeneity must also be taken into consideration when testing for PD-L1, as some areas of the tumor may be PD-L1 positive and others PD-L1 negative. In fact, this may explain why there are patients who we call PD-L1 negative who respond to these agents because those patients are actually PD-L1 positive.

How are you using PD-L1 testing for your patients with NSCLC? Leave a comment in the space below.

Continue

Poll

1.
In your practice, how are you using PD-L1 testing for your patients with metastatic lung cancer?
Submit