Pembrolizumab added to chemotherapy for metastatic nonsquamous non-small cell lung cancer increases 1-year survival rates

EBM Focus - Volume 13, Issue 18

Read the full EBM Focus and earn CME credit.

Reference: KEYNOTE-189 Trial (N Engl J Med 2018 Apr 16 early online) (level 1 [likely reliable] evidence)

  • Pembrolizumab is recommended as first-line monotherapy for patients with advanced non-small cell lung cancer who do not have EGFR or ALK gene mutations if tumor PD-L1 expression levels are high.
  • For patients with a range of tumor PD-L1 expression levels, the KEYNOTE-189 randomized trial compared the addition of pembrolizumab vs. placebo to chemotherapy.
  • Pembrolizumab increased estimated 1-year overall survival rates (in 69.2% vs. 49.4% with placebo, p < 0.001), with increased rates regardless of tumor PD-L1 expression levels.

Non-small cell lung cancer (NSCLC) tumor cells can evade immune system attack if they express PD-L1 surface proteins, which suppress the immune response when bound to PD-1 receptors of T cells. This suppression is thought to be counteracted by immunotherapies such as pembrolizumab, a monoclonal antibody that blocks the PD-1 receptor (Ann Transl Med 2018). For patients with advanced NSCLC who do not have EGFR or ALK gene mutations (for which targeted therapies may be appropriate), pembrolizumab is recommended as first-line monotherapy for tumors with high (≥ 50%) PD-L1 expression (NCCN 2018, NEJM 2016).

For patients with metastatic nonsquamous NSCLC with a range of tumor PD-L1 expression levels, the recent KEYNOTE-189 trial evaluated the addition of pembrolizumab to chemotherapy. In this phase III trial, 616 adults were randomized to pembrolizumab 200 mg IV every 3 weeks for up to 35 cycles vs. placebo, and all had chemotherapy with pemetrexed for all cycles with either carboplatin or cisplatin for the first 4 cycles. Exclusion criteria included previous treatment for metastatic disease, EGFR or ALK gene mutations, active autoimmune disease, treatment-related immune impairment, and symptomatic central nervous system metastases. Tumor PD-L1 expression was < 1% in 31% of patients, 1%-49% in 30%, ≥ 50% in 33%, and unknown in the remaining patients. Patients in the placebo group with radiographic progression could cross over to the pembrolizumab regimen. Intention-to-treat analyses were used. The results reported here are from the first planned interim analysis at a median follow-up of 10.5 months.

Pembrolizumab had a beneficial effect on several important outcomes: estimated 1-year overall survival rates were 69.2% with pembrolizumab vs. 49.4% with placebo; progression-free survival rates were 34.1% vs. 17.3%; and rates of tumor response were 47.6% vs. 18.9% (p < 0.001 for each comparison). The increased overall survival persisted in all subgroup analyses, including subgroups stratified by tumor PD-L1 expression levels. Rates of severe adverse events were similar between groups (67%), but rates of severe immune-related adverse events were 22.7% with pembrolizumab vs. 11.9% with placebo (most commonly hypothyroidism and pneumonitis) and treatment discontinuation due to adverse events occurred in 13.8% vs. 7.9% (statistical comparisons were not reported).

Pembrolizumab added to chemotherapy increased overall survival, progression-free survival, and tumor response in patients with metastatic nonsquamous NSCLC without EGFR or ALK gene mutations, with benefits in overall survival regardless of tumor PD-L1 expression levels. Because this is an interim analysis, longer follow-up of these patients will clarify the effect and provide additional information concerning adverse events secondary to immunomodulation. The present findings demonstrate that pembrolizumab added to chemotherapy is beneficial for patients with metastatic nonsquamous NSCLC regardless of PD-L1 levels. Regarding the recommendation that patients with high PD-L1 levels have first-line pembrolizumab monotherapy (NCCN 2018), it is an open question whether or not the addition of chemotherapy may further increase survival. Additional research is still needed to help determine the preferred treatment regimens based on specific patient characteristics.

For more information, see the topic Immunotherapy for non-small cell lung cancer in DynaMed Plus. DynaMed users click here.