NGS Testing

The treatment of advanced non-small cell lung cancer (NSCLC) has evolved exponentially over the past decade with the expanded identification of novel targetable driver mutations, which have directly improved population-level mortality for advanced NSCLC.¹ It is therefore critical that ALL patients with newly diagnosed advanced NSCLC undergo next-generation molecular testing to evaluate for actionable driver alterations. This is particularly important for patients with newly diagnosed lung adenocarcinoma, in whom the bulk of driver mutations occur. That said, driver alterations have been identified in patients with squamous histology, and therefore next-generation sequencing (NGS) should also be considered for these patients, particularly those who are young with minimal smoking history.

It is our recommendation that, whenever possible, tissue NGS testing and blood circulating tumor DNA (ctDNA) testing should occur concurrently. We recommend tissue NGS testing over individual mutation testing due to the improved sensitivity of this modality. In a study of lung adenocarcinoma patients with light/never smoking history and a “negative” result for actionable driver alterations via non-NGS methods, hybridization-capture NGS testing of 287 cancer-related genes identified actionable driver alterations with NCCN-recommended therapies in 26% of patients, and clinical trial options in an additional 39%.² Furthermore, initial broad NGS testing has been shown to be more cost-effective than sequential targeted testing followed by broad NGS testing, with an earlier time to treatment.³ For tissue testing, we also strongly recommend that an NGS platform incorporating RNA sequencing in addition to DNA sequencing be pursued, as RNA sequencing has improved sensitivity to detect actionable alterations, particularly particular gene fusions. In a study of 275 patients without driver alterations on the MSK-IMPACT DNA sequencing platform, a previously undetected alteration was seen in 14% of cases of which METexon 14 skipping mutations, ROS1 fusions, NRG1 fusions, ALK fusions, and RET fusions were most common.⁴Platforms that currently incorporate RNA sequencing include Caris MI Profile testing, FoundationOne Heme test, and Tempus among others.

Incorporating ctDNA testing is important and should ideally be done concurrent with tissue NGS testing. This is because in 20% of cases ctDNA testing will identify alterations not found with tissue testing and visa-versa.⁵Furthermore, ctDNA testing is highly sensitive. In a study of 93 patients with advanced lung adenocarcinoma and insufficient tumor tissue for tissue-based genotyping, detectable ctDNA was found in 89% of patients.⁶ ctDNA testing with a liquid biopsy is also routinely quicker than tumor testing, with a median turnaround time of 9 vs 15 days.⁵ Liquid NGS platforms include Guardant, FoundationOneLiquid CDx, and Inivata, among others.

Understandably, there are scenarios in which concurrent testing is not feasible or practical from a patient or economic perspective. In this case, the order of testing is dependent on the clinical scenario as follows:⁷

Adequate tumor tissue for NGS testing

In this case one can consider upfront tumor tissue testing with the understanding that, in the event of a negative result, liquid biopsy should be pursued. It is also important to keep in mind that bone samples are not ideal for NGS testing because the de-calcification process can destroy DNA and RNA needed for NGS testing, so ample bone tissue should not be considered “adequate”.

Poor Quality Tumor Tissue for NGS testing

In the case of poor quality tumor tissue such as a specimen with few tumor cells or bone specimen, it can be useful to further sub-divide the scenario by tumor burden. In cases with high tumor burden, upfront liquid biopsy should be considered as the likelihood of detecting circulating cell-free tumor DNA is high. In the event that tumor burden is low, upfront tumor tissue testing can be considered with the understanding that, in the event of a negative result, a repeat tumor biopsy and/or liquid biopsy testing will need to be pursued.

No tumor tissue testing for NGS

In the event that no tumor tissue is available for NGS testing, a liquid biopsy should be pursued and, ideally, a repeat tumor biopsy should be scheduled so that a biopsy can be pursued expediently in the event of a negative ctDNA result. It is important to keep in mind that tumor tissue will also be needed for PD-L1 testing in the event no actionable driver alteration is identified and an immunotherapy-based strategy will be pursued.

One other important consideration is the scenario in which one feels they cannot wait for NGS results before starting therapy. The full molecular profile can take on average 2-3 weeks to return if both tumor tissue and blood testing is needed, and there are scenarios in which a patient doesn’t have this time. Some examples may include a patient with high tumor burden who is symptomatic with evidence of organ compromise or the threat of impending organ compromise. In this case, initiation of chemotherapy WITHOUT immunotherapy can be considered. It is critical that immunotherapy is not initiated before the results of NGS testing are obtained. This is because patients with actionable driver alterations tend not to respond well to immunotherapy and can develop life-threatening toxicities if targeted therapies are begun after immunotherapy.⁸ This is important irrespective of other biomarkers of immunotherapy response, such as PD-L1.