Value of Biomarkers for Immunotherapy in Advanced NSCLC Continue to Evolve

Traditional predictive biomarkers for the efficacy of peri-operative immunotherapy for patients with advanced non–small cell lung cancer (NSCLC), such as PD-L1 expression, still hold value but newer biomarker candidates such as minimal residual disease (MRD) are starting to make an impact, according to Solange Peters, MD, PhD. Peters provided an overview of immunotherapy biomarkers in a presentation during the 23rd Annual International Lung Cancer Congress^®.¹

PD-L1 and Others Display Mixed Results

In the phase 3 IMpower010 trial (NCT02486718), investigators enrolled 1280 patients with stage IB-stage IIIA NSCLC following resection and adjuvant chemotherapy. Patients received four 21-day cycles of cisplatin-based chemotherapy, plus pemetrexed, docetaxel, gemcitabine, or vinorelbine. Of the enrolled patients, 1005 were randomly assigned to receive either the anti–PD-L1 agent atezolizumab (Tecentriq) or best supportive care.²

In a subgroup analysis of disease-free survival (DFS) by PD-L1 status, investigators found that patients with high and intermediate PD-L1 levels derived greater benefit from treatment with atezolizumab compared with those with lower levels. The median DFS strongly favored the atezolizumab arm among the 229 patients with PD-L1 expression in at least 50% of tumor cells (HR, 0.43; 95% CI, 0.27-0.68).

Patients with PD-L1 expression in 1% to 49% of tumor cells (n = 247) also experienced a moderate benefit in terms of median DFS (HR, 0.87; 95% CI, 0.60-1.26) when treated with atezolizumab. Lower levels of PD-L1 expression (< 1% of tumor cells) did not confer a significant DFS benefit when the patient was treated with atezolizumab (HR, 0.97; 95% CI, 0.72-1.31).

“[In terms of] PD-L1, high [expression] and some intermediate [expression] drive the benefit,” said Peters, who is the president of the European Society for Medical Oncology and a professor of medical oncology and thoracic malignancies in the Department of Oncology of Lausanne University in Switzerland. “But remember that in the IMpower010 trial PD-L1 is not properly stratified.”

Peters also touched on the role of tobacco history and chemotherapy as potential biomarkers in the IMpower010 trial.

Patients who identified as never smokers (n = 196) did not experience benefit in terms of DFS from treatment with atezolizumab compared with best supportive care (HR, 1.13; 95% CI, 0.77-1.67). Similarly, patients who were current smokers (n = 139) did not experience a significant difference in DFS benefit between the 2 arms (HR, 1.01; 95% CI, 0.58-1.75). Median DFS did favor the atezolizumab arm among patients who were previous smokers (HR, 0.62; 95% CI, 0.47-0.81).

“Remember that smoking history is not reliable in most medical charts,” Peters noted. “It has not been stratified for smoking history, you can have oncogene addiction in current smokers, and you have a very wide confidence interval. So, I do not think it has to be used as a biomarker.”

In terms of chemotherapy regimen, patients treated with doxacetaxel (n = 152) and vinorelbine (n = 303) both derived a DFS benefit from atezolizumab treatment (HR, 0.74; 95% CI, 0.45-1.23 and HR, 0.63; 95% CI, 0.43-0.91, respectively). However, neither gemcitabine (n = 165) nor pemetrexed (n = 385) favored atezolizumab in terms of DFS (HR, 1.03; 95% CI, 0.62-1.70 and HR, 0.91; 95% CI, 0.67-1.24, respectively).

“Is there any reason why gemcitabine will not be a good component to be given with immunotherapy?” Peters asked. “We have some data in advanced disease, but remember it is sequential. It just does not make sense to me; I think it is just a surrogate biomarker of some other selection process for gemcitabine.”

The phase 3 KEYNOTE-091 trial (NCT02504372) enrolled patients with completely resected stage IB, II, or IIIA NSCLC and randomly assigned them to receive either the anti–PD-1 antibody pembrolizumab (Keytruda) or placebo. Peters focused on the biomarker potential of disease stage, treatment with adjuvant chemotherapy, and finished by again reviewing PD-L1.³

Results from a subgroup analysis showed that patients across disease stages derived a DFS benefit with pembrolizumab: stage IB (HR, 0.76; 95% CI, 0.43-1.37), stage II (HR, 0.70; 95% CI, 0.55-0.91), and stage IIIA (HR, 0.92; 95% CI, 0.69-1.24). However, she noted that these results do not make disease stage a sensible stratification tool because it is contrary to other data sets.

Additionally, patients who received adjuvant chemotherapy also benefited from treatment with pembrolizumab in terms of DFS (HR, 0.73; 95% CI, 0.60-0.89). Patients who did not receive adjuvant chemotherapy did not favor the pembrolizumab arm in terms of DFS (HR, 1.25; 95% CI, 0.76-2.05).

“[I do not believe] that giving chemotherapy before immunotherapy can create a vaccine effect,” Peters said. “Does it mean that if you do not receive chemotherapy, you do not receive the vaccine [effect]? No. Patients who do not receive chemotherapy are a certain and defined subset of patients [for] whom it will probably be very difficult to derive benefit from anything you could do afterwards because of fragility, compliance, or exposure to drugs. This is not a biomarker to me.”

Results by PD-L1 tumor proportion score all demonstrated the benefit of pembrolizumab with the most pronounced among those with 1%-49%: less than 1% (HR, 0.78; 95% CI, 0.58-1.03), 1% to 49% (HR, 0.67; 95% CI, 0.48-0.92), and at least 50% (HR, 0.82; 95% CI, 0.57-1.18). Peters noted that although the findings were properly stratified, the patients with lower PD-L1 levels benefitting more than those with high levels did not really make sense.

In the phase 3 CheckMate 816 trial (NCT02998528), patients with stage IB to IIIA resectable NSCLC were randomly assigned to nivolumab (Opdivo) plus platinum-based chemotherapy or platinum-based chemotherapy alone, followed by resection. For this trial, Peters focused on PD-L1 status and chemotherapy type as potential biomarkers.⁴

In terms of event-free survival (EFS), patients with all levels of PD-L1 expression (< 1%; ≥ 1%; 1%-49%; and ≥ 50%) derived a benefit with nivolumab plus chemotherapy. Patients with the highest level of PD-L1 experienced a 76% reduction in the risk of experiencing an EFS event and those with the lowest level reduced their risk by 15%. Peters noted that this was not surprising and that the findings were properly stratified.

When patients received carboplatin-based chemotherapy, they experienced a greater reduction in the risk of experiencing an EFS event compared with those who were treated with cisplatin (69% vs 29%, respectively). Peters downplayed the significance of this finding, saying that carboplatin had not previously been shown to be superior to cisplatin and that these were smaller subgroups with large confidence intervals.

Overall, Peters concluded that PD-L1 expression is proportionally correlated to the magnitude of benefit from immunotherapy in early-stage disease and that the predictive value of the marker is weakened when chemotherapy and immunotherapy combinations are use. She also added the tumor mutational burden is not predictive when an immunotherapy/chemotherapy combination is used and that other subgroups like histology and gender have discrepant results across trials.

MRD and ctDNA Poised for Larger Role

After shifting the focus of the presentation to the more recently studied biomarkers of minimal residual disease (MRD) and ctDNA, Peters noted that data from multiple trials of patients with various stages of NSCLC have demonstrated the strong prognostic power of the markers among patients with localized disease. Peters also outlined that there are now many different reliable assays for MRD detection targeting ctDNA, including Signatear, Archer PCM, and LUNAR assays.

“Keep in mind that there are many discriminants,” Peters said. “You need to have enough plasma and you need to have a minimal detectable frequency. Technique is very important. Of course, histology, size, and even PET-positivity might make a tumor more or less detectable via ctDNA.”

Again, referencing the IMpower010 trial, Peters explained that ctDNA positivity was strongly prognostic. Among those with ctDNA-positive disease the median DFS was 19.1 months for patients who received atezolizumab compared with 7.9 months for those who received best supportive care (HR, 0.61; 95% CI, 0.39-0.94).

Further, the median DFS favoring atezolizumab in both the ctDNA-negative and ctDNA-positive patients. The median DFS was not reached in the atezolizumab arm compared with 31.4 months in the best supportive care arm (HR, 0.69; 95% CI, 0.53-0.89).²

Peters added that she believes describing the tumor microenvironment has the potential to serve as a biomarker in terms of immunotherapy in the NSCLC space in the future.

“One of the techniques [we have is] imaging mass cytometric,” Peters said. “This uses dozens of metal isotope reporter-labeled antibodies to stain all the cells of interest in a manner to increase sample through output. But you could also use a multispectral immunofluorescence or multispectral immunohistochemistry to do that. The idea is to try to understand what is happening when you give chemotherapy plus immunotherapy to a tumor.”

References

1. Peters S. What other biomarkers are relevant for peri-operative immunotherapy? Presented at: 23rd Annual International Lung Cancer Congress®; July 28-30, 2022; Huntington Beach, CA.
2. Wakelee H, Altorki NK, Zhou C, et al. IMpower010: primary results of a phase III global study of atezolizumab versus best supportive care after adjuvant chemotherapy in resected stage IB-IIIA non-small cell lung cancer (NSCLC). J Clin Oncol. 2021;39(suppl 15):8500. doi:10.1200/JCO.2021.39.15_suppl.8500
3. Paz-Ares L, O’Brien MER, Mauer A, et al. Pembrolizumab (pembro) versus placebo for early-stage non-small cell lung cancer (NSCLC) following complete resection and adjuvant chemotherapy (chemo) when indicated: randomized, triple-blind, phase III EORTC-1416-LCG/ETOP 8-15 – PEARLS/KEYNOTE-091 study. Ann Oncol.2022;33(4):P451-453. doi:10.1016/j.annonc.2022.02.224
4. Forde PM, Spicer J, Lu S, et al; CheckMate 816 Investigators. Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer. N Engl J Med. 2022;386(21):1973-1985. doi:10.1056/NEJMoa2202170