Feature

Blood-based assays ‘where we are headed’ to guide therapy decisions

Charu Aggarwal
Charu Aggarwal

Tumor mutational burden seems to have arrived at an uncertain juncture in its exploration as a potential biomarker for response to immunotherapy.

Although tissue tumor mutational burden (TMB) had shown promise as an indicator of response to checkpoint inhibitors, results of exploratory analyses of the KEYNOTE-189 and KEYNOTE-021 trials presented at this year’s International Association for the Study of Lung Cancer World Conference on Lung Cancer showed it had limited utility for predicting response to chemoimmunotherapy.

“These two clinical trials evaluated the use of chemoimmunotherapy in metastatic non-small cell lung cancer,” Charu Aggarwal, MD, MPH, Leslye M. Heisler assistant professor of medicine in the hematology-oncology division at Perelman School of Medicine at University of Pennsylvania, and a HemOnc Today Editorial Board Member, said in an interview with HemOnc Today. “They showed that tissue TMB did not have any significant implications as a biomarker for response to chemoimmunotherapy.”

Aggarwal discussed the potential role of blood-based TMB in predicting response to chemoimmunotherapy, and said she thinks that a composite score based on gene expression profile, PD-L1 expression and TMB may be used in the future to better guide treatment decisions.

Overcoming tissue heterogeneity

The exploratory analyses of KEYNOTE-189 and KEYNOTE-021 evaluated tissue TMB, assessed by whole-exome sequencing of tumor and matched normal DNA, as a biomarker for response to chemoimmunotherapy.

According to Aggarwal, whole-exome sequencing is often difficult to perform.

“If you look at clinical trials, only about 30% to 40% of people have evaluable tissue TMB data,” she said. “The tissue requirements are high, and it’s usually not easily reportable.”

Aggarwal said intratumor heterogeneity may also make this approach less reliable.

“If you’re biopsying one particular area, you’re only getting a partial picture with the tissue biopsy,” she said. “However, because circulating DNA is being shed from all parts of the tumor, we can get a more comprehensive picture with a blood-based assay.”

A new approach to measuring TMB

A retrospective study by Gandara and colleagues published in Nature Medicine reviewed two large randomized trials, the phase 3 OAK trial and the POPLAR trial, to assess “a novel, technically robust, blood-based assay to measure tumor mutational burden,” according to the study researchers.

The results showed that blood TMB identified patients with NSCLC who would experience a clinically significant PFS benefit with the anti-PD-L1 agent atezolizumab (Tecentriq, Genentech) vs. docetaxel in the second-line or higher setting.

Aggarwal also discussed the ongoing BFIRST clinical trial.

“BFIRST is an ongoing clinical trial that is using blood-based genomic testing to guide therapies,” Aggarwal said. “Blood TMB, in contrast to tissue TMB, is easier to attain. I think this is where we are headed, to be able to use blood-based assays to guide therapy.”

A ‘differential effect’

Aggarwal cited a study she and her colleagues conducted to evaluate associations between blood-based TMB and responses to pembrolizumab (Keytruda, Merck), either as monotherapy or with platinum-based chemotherapy, among 66 patients with metastatic NSCLC.

Thirty-one patients (47%), all of whom had a PD-L1 tumor proportion score of 50% or higher, received pembrolizumab monotherapy. The remaining 35 patients received pembrolizumab plus chemotherapy.

Before treatment initiation, researchers obtained plasma from all patients.

They found that patients who received pembrolizumab monotherapy had median OS of 14.8 months. Median OS for patients treated with pembrolizumab and chemotherapy was not reached.

Of the 52 patients with available data (26 patients in each treatment arm), median blood-based TMB was 16.8 mutations/Mb (range, 1.9-52.5). Patients with high TMB, based on a threshold of 16 or greater mutations/Mb, attained longer median PFS than those with lower TMB (13.8 months vs. 4.7 months; HR = 0.27; 95% CI, 0.13-0.55).

The researchers found that loss-of-function mutations in STK11, KEAP and PTEN, and ERBB2 exon 20 insertion mutations, were predictors of poor outcomes. They concluded that blood TMB may be a predictor of outcomes after pembrolizumab-based therapy among patients with metastatic NSCLC.

“This was a very small study and needs to be validated, but it did look like maybe there is a differential effect,” Aggarwal said. “Ours is the only study so far that has looked at blood-based TMB in chemoimmunotherapy, so the future of blood-based TMB remains to be seen.”

According to Aggarwal, a composite biomarker consisting of gene expression profiling, TMB and PD-L1 expression may be the most effective tool to predict outcomes with checkpoint inhibitors.

“Gene expression profiling is done on tissue. There’s some emerging data that shows it might be helpful in guiding therapy,” she said. “I think no one biomarker is going to be perfect. We need to use a combination of different biomarkers to guide therapy.”– by Jennifer Byrne

References:

Aggarwal C, et al. Abstract MA25.04. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona, Spain.

Gadgeel S, et al. Abstract LBA81_PR. Presented at: European Society for Medical Oncology Congress; Sept. 27-Oct. 1, 2019; Barcelona, Spain.

Gandara DR, et al. Nat Med. 2018;doi:10.1038/s41591-018-0134-3.

Garassino MC, et al. Abstract OA04.06. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona, Spain.

Langer C, et al. Abstract OA04.05. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona, Spain.

For more information:

Charu Aggarwal, MD, MPH, can be reached at 3400 Spruce St., Philadelphia, PA 19104; email: charu.aggarwal@pennmedicine.upenn.edu.

Disclosures: Aggarwal reports advisory board roles with AstraZeneca, Bristol-Myers Squibb, Celgene, Merck and Roche, and research funding to her institution from AstraZeneca, Incyte, MacroGenics and Merck.

Charu Aggarwal
Charu Aggarwal

Tumor mutational burden seems to have arrived at an uncertain juncture in its exploration as a potential biomarker for response to immunotherapy.

Although tissue tumor mutational burden (TMB) had shown promise as an indicator of response to checkpoint inhibitors, results of exploratory analyses of the KEYNOTE-189 and KEYNOTE-021 trials presented at this year’s International Association for the Study of Lung Cancer World Conference on Lung Cancer showed it had limited utility for predicting response to chemoimmunotherapy.

“These two clinical trials evaluated the use of chemoimmunotherapy in metastatic non-small cell lung cancer,” Charu Aggarwal, MD, MPH, Leslye M. Heisler assistant professor of medicine in the hematology-oncology division at Perelman School of Medicine at University of Pennsylvania, and a HemOnc Today Editorial Board Member, said in an interview with HemOnc Today. “They showed that tissue TMB did not have any significant implications as a biomarker for response to chemoimmunotherapy.”

Aggarwal discussed the potential role of blood-based TMB in predicting response to chemoimmunotherapy, and said she thinks that a composite score based on gene expression profile, PD-L1 expression and TMB may be used in the future to better guide treatment decisions.

Overcoming tissue heterogeneity

The exploratory analyses of KEYNOTE-189 and KEYNOTE-021 evaluated tissue TMB, assessed by whole-exome sequencing of tumor and matched normal DNA, as a biomarker for response to chemoimmunotherapy.

According to Aggarwal, whole-exome sequencing is often difficult to perform.

“If you look at clinical trials, only about 30% to 40% of people have evaluable tissue TMB data,” she said. “The tissue requirements are high, and it’s usually not easily reportable.”

Aggarwal said intratumor heterogeneity may also make this approach less reliable.

“If you’re biopsying one particular area, you’re only getting a partial picture with the tissue biopsy,” she said. “However, because circulating DNA is being shed from all parts of the tumor, we can get a more comprehensive picture with a blood-based assay.”

A new approach to measuring TMB

A retrospective study by Gandara and colleagues published in Nature Medicine reviewed two large randomized trials, the phase 3 OAK trial and the POPLAR trial, to assess “a novel, technically robust, blood-based assay to measure tumor mutational burden,” according to the study researchers.

The results showed that blood TMB identified patients with NSCLC who would experience a clinically significant PFS benefit with the anti-PD-L1 agent atezolizumab (Tecentriq, Genentech) vs. docetaxel in the second-line or higher setting.

Aggarwal also discussed the ongoing BFIRST clinical trial.

“BFIRST is an ongoing clinical trial that is using blood-based genomic testing to guide therapies,” Aggarwal said. “Blood TMB, in contrast to tissue TMB, is easier to attain. I think this is where we are headed, to be able to use blood-based assays to guide therapy.”

PAGE BREAK

A ‘differential effect’

Aggarwal cited a study she and her colleagues conducted to evaluate associations between blood-based TMB and responses to pembrolizumab (Keytruda, Merck), either as monotherapy or with platinum-based chemotherapy, among 66 patients with metastatic NSCLC.

Thirty-one patients (47%), all of whom had a PD-L1 tumor proportion score of 50% or higher, received pembrolizumab monotherapy. The remaining 35 patients received pembrolizumab plus chemotherapy.

Before treatment initiation, researchers obtained plasma from all patients.

They found that patients who received pembrolizumab monotherapy had median OS of 14.8 months. Median OS for patients treated with pembrolizumab and chemotherapy was not reached.

Of the 52 patients with available data (26 patients in each treatment arm), median blood-based TMB was 16.8 mutations/Mb (range, 1.9-52.5). Patients with high TMB, based on a threshold of 16 or greater mutations/Mb, attained longer median PFS than those with lower TMB (13.8 months vs. 4.7 months; HR = 0.27; 95% CI, 0.13-0.55).

The researchers found that loss-of-function mutations in STK11, KEAP and PTEN, and ERBB2 exon 20 insertion mutations, were predictors of poor outcomes. They concluded that blood TMB may be a predictor of outcomes after pembrolizumab-based therapy among patients with metastatic NSCLC.

“This was a very small study and needs to be validated, but it did look like maybe there is a differential effect,” Aggarwal said. “Ours is the only study so far that has looked at blood-based TMB in chemoimmunotherapy, so the future of blood-based TMB remains to be seen.”

According to Aggarwal, a composite biomarker consisting of gene expression profiling, TMB and PD-L1 expression may be the most effective tool to predict outcomes with checkpoint inhibitors.

“Gene expression profiling is done on tissue. There’s some emerging data that shows it might be helpful in guiding therapy,” she said. “I think no one biomarker is going to be perfect. We need to use a combination of different biomarkers to guide therapy.”– by Jennifer Byrne

References:

Aggarwal C, et al. Abstract MA25.04. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona, Spain.

Gadgeel S, et al. Abstract LBA81_PR. Presented at: European Society for Medical Oncology Congress; Sept. 27-Oct. 1, 2019; Barcelona, Spain.

Gandara DR, et al. Nat Med. 2018;doi:10.1038/s41591-018-0134-3.

Garassino MC, et al. Abstract OA04.06. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona, Spain.

Langer C, et al. Abstract OA04.05. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona, Spain.

For more information:

Charu Aggarwal, MD, MPH, can be reached at 3400 Spruce St., Philadelphia, PA 19104; email: charu.aggarwal@pennmedicine.upenn.edu.

Disclosures: Aggarwal reports advisory board roles with AstraZeneca, Bristol-Myers Squibb, Celgene, Merck and Roche, and research funding to her institution from AstraZeneca, Incyte, MacroGenics and Merck.

    See more from HemOnc Today's PharmAnalysis