Clinical Decisions in Non-Driver NSCLC - Episode 4

Second-Line Immunotherapy in Advanced Non-Driver NSCLC

April 10, 2018

Benjamin P. Levy, MD:We now have 3 drug approvals for second-line non—small cell lung cancer, with immunotherapy. We’ve got single-agent pembrolizumab, single-agent nivolumab, and single-agent atezolizumab, all showing superior outcomes when compared with docetaxel. Pembrolizumab and nivolumab are PD-1 drugs. Atezolizumab is a PD-L1 drug. If you look at cross-trial comparisons between KEYNOTE-010, CheckMate-017 and CheckMate-057, and the OAK trial, these drugs elicit similar response rates and similar outcomes overall. So, it’s hard for me to really state that one of these drugs is better than the other. Numerically, the longest overall survival we’ve had from an immunotherapy has been from the OAK trial, and that was 13.2 months, which exceeds the survival we saw with pembrolizumab and nivolumab in the KEYNOTE and CheckMate trials, respectively.

That said, I think it’s really dealer’s choice and what you feel familiar with. Nivolumab has now moved to be an every-4-week regimen. Pembrolizumab and atezolizumab as second-line drugs are every 3 weeks. It’s just what you’re comfortable with. There’s some biological rationale why atezolizumab, a PD-L1 drug, may have less toxicity. But I can tell you, in my experience, the toxicities between these 3 agents are fairly similar.

Assessing a response in a patient with immunotherapy can be a challenge. There have been other immune-related response criteria that have been looked at as well. It’s sometimes difficult because these drugs are not cytotoxic and they’re not going to elicit responses similar to what we see in something like chemotherapy and antiangiogenic strategies. There is pseudoprogression that we need to factor in here with these drugs. The rate of pseudoprogression is low, and there are hyper-progressers, too—people who receive these drugs who progress quite rapidly. But I think we need to be mindful that there may be changes within the tumor, elicited by these drugs, that don’t necessarily mean the patient is progressing. It may have to do with the fact that there’s more T-cell trafficking in an inflamed tumor that may show something on a scan that looks bigger. But nevertheless, the patient is deriving a benefit. I think we need to be very diligent when we’re looking at the responses. I still image patients every 4 to 6 treatments, or every 3 months, on these drugs. But I’m very careful when I look at the images when making a decision to move forward.

Transcript edited for clarity.


  • A 55-year old female presented with chronic cough and 10-lb weight loss
  • PMH: never smoker; no family history of cancer; no known exposure to chemicals or asbestos
  • Chest x-ray showed a 5.0-cm lesion in the left lower lobe with bulky lymphadenopathy
  • Chest CT scan confirmed the presence of a lung mass and enlargement of the right hilar lymph node and bilateral mediastinal lymph nodes
  • EUS-guided biopsy was performed
  • Pathology revealed adenocarcinoma
  • Molecular testing:
    • FISH: negative for ALK translocation
    • NGS: negative for EGFR, ROS1, RET, BRAF, KRAS
    • IHC: PD-L1 expression in 0% of cells
  • PET/CT imaging showed 18F-FDG uptake in the lung mass, right hilar lymph node, mediastinum, and left adrenal gland
  • MRI of the brain was normal
  • ECOG PS, 0
  • The patient was started on therapy with carboplatin/pemetrexed and bevacizumab
  • The regimen was well tolerated
  • After 6 cycles, the patient had a good response
  • She was continued on bevacizumab
  • After 9 months on therapy, the patient developed cough and weight loss
  • Follow-up imaging revealed multiple new lesions in the left adrenal gland and new liver metastases
  • Patient was started on atezolizumab, planned for 12 months