Management and Treatment of Patients With Exon 20 Non–Small Cell Lung Cancer

Latest Recommendations for Broad Molecular Testing in Advanced NSCLC

During a live virtual event, Zofia Piotrowska, MD, MHS, discussed the National Comprehensive Cancer Network guidelines for molecular testing for a patient with advanced non–small cell lung cancer, and how to approach a patient who is found to have an EGFR exon 20 insertion. This is the first of 2 articles based on this event.


  • A 56-year-old woman presented with persistent cough and mild progressive dyspnea over last month, with some chest discomfort, especially during routine exercise​.
  • Mother of 2 teenage children, former light smoker, no other notable medical history​
  • ECOG performance score: 1​
  • CT/PET-CT showed mass in right lower lobe, right pleural metastasis, and right malignant pleural effusion​.
  • Brain MRI was negative for brain metastases​.
  • Transbronchial lung biopsy showed adenocarcinoma.​
  • Tissue NGS showed EGFR exon 20 insertion mutation (V769_D770insASV)​.
  • PD-L1 expression: negative​

Targeted OncologyTM: What do the latest National Comprehensive Cancer Network (NCCN) guidelines recommend for biomarker testing for patients with advanced or metastatic non–small cell lung cancer (NSCLC)?

ZOFIA PIOTROWSKA, MD, MHS: For patients with advanced or metastatic NSCLC, obviously we need the tissue to establish a diagnosis and then molecular testing is recommended. [You can] look for all these different targets that we now have approved therapies for, including EGFR, ALK, KRAS, ROS, BRAF, NTRK, MET exon 14 skipping, and RET.1 The NCCN does make the point about broad molecular profiling as well as PD-L1 testing. All of this is recommended for the adenocarcinomas, large cell histology, and the non-small cell NOS [type not otherwise specified], and certainly to be considered in patients with squamous cell carcinomas, particularly those who may have a limited smoking history or other factors that make you think they’re more likely to have a driver.

Does broad molecular testing need to include all potentially actionable genes even if this takes longer to get results? What about testing RNA versus DNA?

In my experience, when they say broad molecular testing, the most important are the ones listed [by the NCCN] which are the targets with the approved agents. Then the second-tier group, which may have clinical trials like HER2, non-G12C KRAS mutations, and others. I personally think that the [Guardant 360 83-gene] panel covers pretty much everything that’s clinically relevant. There may be rare cases where you have a young, never-smoker and there’s nothing on that where you might think about sending a broader panel if the 83-gene panel doesn’t come back with something informative. But I think for most patients the 83-gene panel and the quicker turnaround time seems like a very reasonable compromise. More I don’t think is always better.

I would say all of the rearrangements and fusions are a little bit more easily detected on an RNA-based fusion panel. What many centers, including ours, have done is start with a DNA-based NGS [next-generation sequencing] and that can pick up many of these, but if you don’t find something then moving on to the RNA-based fusion panel. That can increase your sensitivity for the fusions, the MET exon 14 skipping, and those other events. So that’s an important point, too.

[There are many] different targets that we have approved therapies for, including the various types of EGFR mutations that we now look for, as well as KRAS, ALK, ROS1, BRAF V600E, NTRK, MET exon 14 skipping, and RET, as well as considering PD-L1 as a biomarker for patients without a targetable alteration.

To briefly highlight the NCCN recommendations, there is a category 1 recommendation for testing these different targets, ideally as part of broad molecular testing because a broad molecular panel minimizes tissue waste and it helps identify rare driver mutations. Here what we’re saying is a broad-based NGS panel as opposed to single-gene PCR [polymerase chain reaction] test as a standalone, which I think is now inadequate as a way to test NSCLC with all of these targets that we’re looking for. Consider RNA NGS for those patients without an identifiable driver oncogene. If there is insufficient tissue to allow for testing of all of these, it is important, as we were highlighting earlier, to consider either a liquid biopsy or plasma testing.

Why does molecular testing need to be performed prior to systemic therapy?

The guidelines recommend that ideally you should have the molecular testing results before starting first-line therapy, especially first-line immunotherapy. The reason for that is although PD-L1 expression can be elevated in patients with oncogene drivers, for those patients, such as patients with EGFR and ALK, that high PD-L1 doesn’t necessarily translate into high response rates to immunotherapy. The important point here is that if those patients have a targetable alteration, you want to start with a targeted therapy before using immunotherapy because of efficacy, but also because of this concern that if you start them on immunotherapy and then want to pivot to a targeted therapy, like osimertinib [Tagrisso] or another tyrosine kinase inhibitor, there is an increased risk of adverse events [AEs]—specifically pneumonitis with osimertinib and other immune-related AEs, hepatitis with crizotinib [Xalkori], and others. For patients who have a targetable driver oncogene, we want to try to use that first-line targeted therapy rather than immunotherapy even if their PD-L1 score is high.

What treatment approach is recommended for patients with NSCLC who have a confirmed EGFR exon 20 insertion?

The guidelines for patients with EGFR exon 20 insertion mutation-positive NSCLC say first-line therapy should be along the regular guidelines for NSCLC, typically either chemotherapy or chemoimmunotherapy. I generally tend to use chemotherapy for these patients, but I think chemoimmunotherapy is reasonable if you feel that you want the patient to get immunotherapy. And then upon progression, both amivantamab [Rybrevant] and mobocertinib [Exkivity] are approved. There’s no clear guidance as to which one is preferred, or in which order they should be given. They have some similarities and some differences.


1. NCCN. Clinical Practice Guidelines in Oncology. Non–small cell lung cancer, version 4.2022. Accessed September 13, 2022.