Proteomics and Prognostics May Serve as Potential Biomarkers Beyond PD-L1 in Lung Cancer

Luis Raez, MD

Newer biomarkers including the plasma proteomics–based PROphet test may provide the necessary innovation in lung cancer to help determine response to treatment, which is especially important as PD-L1 is a well-known but not entirely accurate biomarker, according to Luis Raez, MD. With the need for more accurate biomarkers, poor prognostic biomarkers have demonstrated promise, but others such as tumor mutation burden (TMB) and liquid biopsies require further investigation.

“We need to tailor immunotherapy [approaches] better because we want something more cost effective and less toxic for patients,” Raez said in an interview with OncLive® regarding data presented at a State of the Science Summit™ on lung cancer, which he chaired. “If we don’t know what markers are good or [poor prognostically], the possibility of using circulating tumor DNA [ctDNA] to try to classify the patients better, so we know who’s going to benefit from immunotherapy, [could be promising].”

In addition to expanding on research with ctDNA in lung cancer, Raez noted that one potential way to tailor treatments for patients with lung cancer receiving PD-(L)1 inhibitor–based therapies could be using proteomics.

Findings from the multicenter observational PROPHETIC trial (NCT04056247) revealed that combining standard PD-L1 testing with plasma proteome–based testing helped identify patient subsets with different outcomes from PD-(L)1–based regimens. PROphet differentiated between patients with non–small cell lung cancer with PD-L1 expression levels of 50% or higher deemed PROphet-negative, who have improved overall survival when treated with immunotherapy plus chemotherapy vs immunotherapy alone (HR, 0.23; 95% CI, 0.1-0.51, P < .0003). The test demonstrated PD-L1–high patients who were PROphet-positive showed comparable outcomes with chemoimmunotherapy and immune monotherapy (HR, 0.78; 95% CI, 0.42-1.44, P = .424).1

In the interview, Raez detailed several potential biomarkers in the lung cancer space under investigation. Raez is medical director and chief scientific officer of Memorial Cancer Institute in Pembroke Pines, Florida.

OncLive: What are notable recent advancements with biomarkers in lung cancer?

Raez: PD-L1 is a well-known biomarker for immunotherapy. It’s a bit frustrating because it’s not 100% accurate. For example, patients with a PD-L1 [expression of] more than 50% [are among] the best [responding] group of patients to immunotherapy. However, even in that group of patients, only approximately 40% respond, so there are patients who have high PD-L1 [expression] and do not respond [to immunotherapy]. It’s frustrating because we realize that the biomarker is not very accurate. That’s why we’re in search of [new] biomarkers.

For years [we have been examining] TMB which has not been out yet [as a biomarker] because there is no harmonization. Everybody calculates that biomarker the way that they want, so it’s hard to [standardize its use]. [Therefore], we don’t have an answer to that [as a potential biomarker] yet.

We also have been working with [poor prognostic] biomarkers; we used to be looking for biomarkers [that predict response]. Now, we know that if the patient has an STK11 mutation [or] KEAP1 mutation, they will not do well with immunotherapy. We cannot get [enough information on] the good [prognostic biomarkers], but at least now we know which tumors don’t do very well. If patients have [poor prognostic] biomarkers like STK11and KEAP1, the phase 3 POSEIDON trial [NCT03164616] regimen that uses anti­–CTLA-4 and anti–PD-L1 inhibitors plus chemotherapy, has evidence that it will benefit patients. When patients have these [poor prognostic] mutations, they don’t experience any benefit from immunotherapy [if] the regimens [include] only anti–PD-1 agents.

What is the potential of a proteomics biomarker in lung cancer?

A new biomarker that’s very interesting [centers around] proteomics, the proteins in the blood. There is a test called PROphet [and] we have been able to identify the proteins that are associated with lack of response. If we find these proteins, we know that the patient is not going to do very well with immunotherapy. For that reason, we can classify patients by who will respond and who will not respond to immunotherapy. In that way, we help the PD-L1 [biomarker] because patients with a high PD-L1 [expression] may respond [to immunotherapy], but now if the PROphet test is positive, we can know with more certainty that the patients are going to do better [with immunotherapy].

This is a very interesting [approach]. For [decades] we have known about proteomics, and we still don’t have a standard of care application in cancer. This may be the first one. We [could] do this blood test that can help us to modulate and improve the response that we see currently with the only [available] marker, PD-L1.

How could liquid biopsies serve as a promising biomarker in lung cancer?

The [other biomarker] we love is liquid biopsies [as] there is ctDNA in the blood. A lot of current studies [show] that ctDNA correlates with the presence of the tumor. If the ctDNA is disappearing in the blood, the patient is getting better. We can use that information to change the course of treatment or to intensify therapy.

For example, if we give a patient immunotherapy alone, and the ctDNA [levels] don’t go down—they can even go up—that patient may also [need] chemotherapy; immunotherapy alone is not going to do the job because the ctDNA is not clearing. But if that ctDNA is clearing, perhaps immunotherapy alone is what the patient needs. That’s why there is a lot [of research] in this area that needs to [be] improved upon.

How could these advancements with biomarker research influence your treatment decision-making process with patients?

It’s frustrating for us because on one side of lung cancer therapy we are [applying] precision medicine with targeted therapy. If a patient has an EGFR mutation, they get EFGR-[targeted] therapy, [and the same with ALK mutations and] ALK-targeted therapy. Immunotherapy seems to be sophisticated, but if you think about giving immunotherapy to every patient not knowing who is going to respond, that is not precision medicine.

It’s the same thing that we were doing 20 years ago with chemotherapy; we gave the same chemotherapy to every patient, and we didn’t know who was going to respond. The only difference now is we’re [administering] a more sophisticated, less toxic [treatment] but at the same time it’s more expensive. For that reason, we need to tailor the immunotherapy and try to identify groups of patients who are going to respond [such as those with] PD-[L]1–high [expression or those who have a] PROphet-positive test; or [we must] identify the groups who are not going to respond [such as those with] STK11 and KEAP1 mutations.

If patients are not going to respond, perhaps they should not receive immunotherapy or should receive only immunotherapy [based on certain factors]. [Consider] the example of the POSEIDON regimen or the phase 3 CheckMate 9LA trial [NCT03215706] regimen—those are the only regimens that have proved beneficial in this population of patients with STK11 and KEAP1 mutations.

Reference

Christopoulos P, Harel M, McGregor K, et al. Plasma proteome–based test for first-line treatment selection in metastatic non–small cell lung cancer. JCO Precis Oncol. 2024;8:e2300555. doi:10.1200/PO.23.00555