PD-L1 Expression Remains Primary Guide for NSCLC Treatment Selection in the Absence of Targetable Mutations

Kaushal Parikh, MBBS

Kaushal Parikh, MBBS, discussed how PD-L1 expression affects treatment decision-making in the absence of other targetable mutations or prognostic indicators in patients with non–small cell lung cancer (NSCLC) in an interview with OncLive^®. He also detailed data supporting the use of quadruplet regimens for patients who have low PD-L1 expression and other poor prognostic indicators.

“Approximately 65% to 70% of patients have some oncogenic driver alterations. Of course, not all of them are targetable,” said Parikh, a thoracic oncologist and assistant professor of oncology at the Mayo Clinic Comprehensive Cancer Center in Rochester, Minnesota. “Therefore, you still have a majority of a population of patients who are not candidates for targeted therapy.”

In a concurrent interview, Parikh also highlighted treatment advances in early-stage NSCLC, including investigations of adjuvant immunotherapy for patients with actionable mutations and questions on how to properly tailor the use of this highly effective approach.

OncLive: What percentage of patients with NSCLC harbor oncogenic driver alterations, and how does this impact the standard of care for those without targetable mutations?

Parikh: Targetable oncogenic drivers [occur in] approximately 40% to 50% of patients with NSCLC, [and] that includes first line [and] second line targets including KRAS G12C. However, in patients [who are not candidates for targeted therapy], immunotherapy either by itself or in combination with another immunotherapeutic agent with or without platinum chemotherapy agents is the standard of care.

We still have only 1 biomarker to [guide] treatment for these patients, which is PD-L1 expression. Tumor mutational burden is not a biomarker that we routinely use in NSCLC. There’s a tumor-agnostic approval for pembrolizumab [Keytruda] for any malignancy with a high tumor mutational burden, but that is not something that we base our treatments on right now in lung cancer.

In the absence of targetable oncogenic drivers, how does PD-L1 expression influence the choice between single-agent checkpoint inhibitors and combination regimens?

For most patients who have high PD-L1 expression of greater than 50%, it’s reasonable to still use a single immune checkpoint inhibitor–based regimen such as pembrolizumab, cemiplimab-rwlc [Libtayo], or atezolizumab [Tecentriq], all 3 of which are approved. The situations in which I would offer chemotherapy to patients with high PD-L1 expression is for those with very high symptomatic disease burden where [they need] a quick response, [for] patients who are not candidates for immunotherapy because of autoimmune diseases such as lupus, or for patients [for whom] pneumonitis can potentially be fatal. For most patients, even those with autoimmune conditions such as rheumatoid arthritis, I would be willing to try immunotherapy since the benefits generally outweighs the risks.

For patients with less than 50% PD-L1 expression, I typically prefer a combination regimen. Single-agent pembrolizumab has been studied for patients with a PD-L1 expression of 1% or greater in the phase 3 KEYNOTE-042 trial [NCT02220894], and it did show an improvement in survival. However, when we look at the subset analyses, most of this benefit was driven by those with high PD-L1 expression. That would not be a regimen I would consider for most patients. Platinum doublet and single-agent [treatment with a checkpoint inhibitor], a triplet therapy, is usually my go-to regimen for patients. [I would use] pemetrexed-based regimens for patients with adenocarcinoma and a taxane-based regimen for those with non-adenocarcinomas or squamous cell cancers in first line treatment.

What are the benefits and current limitations of quadruplet regimens for patients with low PD-L1 expression and poor prognostic indicators, such as KRAS or STK11 mutations?

The interesting nuances here are for patients who have squamous histology, have less than 1% PD-L1 expression, or have poor prognostic indicators. That’s where [treatment selection] gets a little murkier. Although they are not prospective, phase 3 data, we do have some data from subset analyses regarding the role of doublet immune checkpoint inhibition in these patients, both with PD-1/PD-L1 inhibition as well as CTLA-4 blockade.

In the phase 3 CheckMate-227 trial [NCT02477826] of nivolumab [Opdivo] plus ipilimumab [Yervoy], the 5-year overall survival [OS] was significantly better [with the combination] compared with chemotherapy alone. This signal was even stronger for patients who had squamous cell histology and PD-L1 expression of less than 1%. This was not the primary end point of the study, so the regimen is not approved by the FDA for this indication. However, this is a subset of patients who benefit from this immunotherapy combination.

Then we have 2 quadruplet regiments, 3 if we consider the phase 3 IMpower-150 trial [NCT02366143] regimen of atezolizumab [Tecentriq] plus bevacizumab [Avastin] and platinum doublet chemotherapy, which I believe has fallen out of favor. VEGF and PD-L1 inhibition has been of significant interest over the past several years now, but that is not a regimen that I use or that most of my colleagues use in the first-line setting.

The other 2 quadruplet regiments are the phase 3 CheckMate 9LA [NCT03215706] and POSEIDON [NCT03164616] regimens, both of which have a platinum-based doublet with either nivolumab and ipilimumab or durvalumab [Imfinzi] and tremelimumab [Imjudo], [respectively]. Both studies [data] have shown a signal for patients with low PD-L1 expression. Subset analyses of POSEIDON have analyzed [data in] patients with poor prognostic markers such a KRAS or STK11 mutation, as well as a KEEP1 alteration. These mutations typically render the tumor into a cold microenvironment, [and patients then] have [poor] responses to immunotherapy and poor survival overall. The addition of a CTLA-4 inhibitor to [an anti–PD-1 inhibitor and chemotherapy] was shown to improve survival.

We don’t have prospective phase 3 data for these combinations right now, but there are ongoing studies assessing [whether] patients with high-risk disease [harboring] KRAS, STK11, or KEEP1 alterations [experience greater] benefit with a quadruplet regimen over a triplet regimen.

For patients with non-squamous KRAS-mutated lung cancer in the POSEIDON trial, the median OS was 25.7 months with the quadruplet regimen vs 10.4 months with chemotherapy alone. That’s a significant difference in [outcomes for] patients with KRAS mutations compared with the overall study population [where the median OS was 14.0 months vs 11.6 months, respectively]. When we further look at patients with STK11 mutations, the benefit in OS is approximately 5 months with the POSEIDON regimen compared with chemotherapy. With time we’ll find out from prospective studies whether [the data with] this quadruplet regimen holds true.