July 2016 Edition Vol.11, Issue 7

Biomarkers for Immunotherapy: Research at ASCO and Treatment Implication

Biomarkers for Immunotherapy: Research at ASCO and Treatment Implications

By Gena Kanas, M.P.H., Ph.D., Consultant, Clinical & Scientific Assessment, Kantar Health

New data presented at ASCO 2016 showed the clinical implications of biomarkers in evaluating responses to immunotherapy. In this discussion, we will illustrate how potential biomarkers are being investigated and used to evaluate responses to immunotherapy in the treatment of a variety of cancers, including non-small cell lung cancer (NSCLC), melanoma, and urothelial carcinoma.

Epidemiology of Tumors Currently Treated with FDA-Approved Immunotherapy

Three U.S. Food and Drug Administration (FDA)-approved PD-1/-L1 checkpoint inhibitor immunotherapies (Keytruda® [pembrolizumab, Merck & Co.], Opdivo® [nivolumab, Bristol Myers-Squibb/Ono Pharmaceuticals], and Tecentriq™ [atezolizumab, Genentech/Roche]) have shown clinical activity with durable response and prolonged overall survival (OS) in several cancers.

Current FDA approvals include the use of Keytruda or Opdivo in advanced melanoma and advanced NSCLC; Opdivo for use in advanced renal cell carcinoma (RCC) and Hodgkin’s lymphoma; and most recently approved Tecentriq in advanced urothelial carcinoma.

Most of these cancers have low survival at advanced stages (Table 1).

Unfortunately, over half of patients treated with immunotherapy do not show a clinical response, which has led to further investigation into possible immunotherapy combination strategies and to identify potential biomarkers to better predict those patients who best respond to immunotherapy.

*All incidence and survivial figures are sourced from CancerMPact® Patient Metrics. Kantar Health. Available at www.cancermpact.com. Accessed June 16, 2016.

Possible Biomarkers for Immunotherapy

Current immunotherapies target the interaction between PD-1 and PD-L1, and the inhibition of this interaction leads to antitumor activity.

The expression of PD-L1 has been detected in several different tumor and/or immune cells, including those listed in Table 1 as well as in gastric cancer, ovarian cancer, head and neck cancer and others.

The challenge with using PD-L1 expression as a predictive biomarker is that its expression is dynamic within an individual, and a patient may be considered positive at one point in time or with certain cells (such as with metastases vs primary lesion) but negative at another time point or with other cells.1-6

The goal of having biomarkers is to select only those patients who will respond to a given treatment.

Tumor mutation burden and mismatch repair deficiency, in addition to PD-L1 expression, as possible biomarkers were reported at this year’s ASCO conference.

PD-L1 Expression

Two abstracts focused on response to immunotherapy based on PD-L1 expression in patients with NSCLC: KEYNOTE-010 and KEYNOTE-021 studies.

Keytruda’s accelerated FDA approval in platinum-pretreated NSCLC limits utilization to patients with PD-L1-overexpressing tumors.

The confirmatory randomized Phase II/III KEYNOTE-010 study enrolled patients with advanced PD-L1-positive NSCLC (n=1,034) who had received at least one prior chemotherapy. Patients were randomized to Keytruda (2 or 10 mg/kg Q3W) or docetaxel (75 mg/m2 Q3W) for 24 months or loss of benefit.

The authors reported a subgroup analysis from the KEYNOTE-010 study that investigated PD-L1 expression as a biomarker for outcomes.7

Patients included in this subgroup study had to exhibit PD-L1 expression on at least 1% of tumor cells. KEYNOTE-010 had already shown an OS advantage for either dose level of Keytruda compared with docetaxel.8

The authors reported a general correlation for improved efficacy compared with docetaxel with an increasing tumor proportion score (TPS), with those patients with TPS of 50% or higher having the best outcomes.

This correlation held for overall response rate (ORR; TPS 50-74%: 22.6% vs 9.6%, p=0.01 and TPS 75%-100%: 33.7% vs 7.0%, p<0.0001), progression-free survival (PFS; TPS 75%-100%: HR 0.52, p<0.0001) and OS (TPS 50-74%: HR 0.58, p=0.01 and TPS 75%-100%: HR 0.51, p<0.0001).

All patients treated with Keytruda had at least as good of an outcome as those treated with docetaxel. Unfortunately, since the investigators did not include PD-L1-negative patients, KEYNOTE-010 was unable to determine the true utility of PD-L1 status as a biomarker for Keytruda.

The KEYNOTE-021 study9 investigated the combination of Keytruda (2 or 10 mg/kg Q3W) plus platinum doublet chemotherapy for four cycles as first-line therapy followed by Keytruda maintenance among three cohorts of patients (Cohort A [n=25], Cohort B [n=25], Cohort C [n=24]) with advanced NSCLC.

The three cohorts differed by the type of chemotherapy used in combination with Keytruda. The ORR was 56% in all patients, with one complete response. Patients were tested for PD-L1 TPS status and included patients with TPS of less than 1%.

In this analysis, the authors reported improved efficacy of Keytruda combined with chemotherapy regardless of PD-L1 status.

Across all three cohorts, the ORR was similar for both TPS of 1% or higher and TPS of less than 1% (Cohort A: 53% vs 44%; Cohort B: 50% vs 40%; Cohort C: 69% vs 75%; Total: 57% vs 54%, respectively).

The authors also reported data on OS and PFS for the three cohorts; however, these data were not stratified by TPS status.     

These study results show that PD-L1 expression is correlated with clinical response to immunotherapy, although this correlation may be limited to use of these agents as monotherapies.

The problem with PD-L1 expression as a biomarker is that the responses to therapy are greater among those who express PD-L1, but those who are considered PD-L1 negative also respond.

This suggests there are additional factors influencing the response to anti-PD-1/PD-L1 therapy.

Tumor Mutation Burden

Three abstracts focused on tumor mutation burden (TMB) as a possible predictor of immunotherapy response in patients with NSCLC, melanoma, and urothelial carcinoma.

A study from the Sarah Cannon Research Institute investigated the TMB in NSCLC using comprehensive genomic profiling on tumor specimens.10

The authors reported a correlation between TMB (calculated as the number of synonymous and nonsynonymous variants from a series of 236-315 genes) and longer treatment duration with immunotherapy (Keytruda, Opdivo, or avelumab (MSB0010718C, Pfizer)) in 64 NSCLC cases.

It is important to note that this study did not report data on efficacy of treatment with immunotherapy by level of TMB.

Patients with TMB of 15 or more mutations/Mb had a longer treatment duration with immunotherapy compared with those with TMB of less than 15 mutations/Mb (median 64 weeks vs 17 weeks, p=0.010).

When comparing treatment duration between those classified as having high TMB (≥12.1 mutations/Mb) and those with low TMB (≤3.2 mutations/Mb), the authors reported some patients with high TMB had shorter treatment durations, and some patients with low TMB with longer treatment durations.  

Douglas Johnson, M.D., MSCI, Vanderbilt Ingram Cancer Center, presented next-generation sequencing data in 65 melanoma tumor specimens to identify markers of response to anti-PD-1 therapy (Opdivo, Keytruda or Tecentriq).11

TMB was evaluated using a small fraction (236-315 genes, same set of genes analyzed by the previous study) of the genome as a surrogate for whole genome TMB. A high TMB was defined as more than 23.1 mutations/Mb, intermediate defined as 3.3-23 mutations/Mb, and low TMB defined as less than 3.3 mutations/Mb.

Dr. Johnson reported a correlation between increasing TMB and higher PFS (high: not reached vs intermediate: 89 days vs low: 86 days, p<0.001) and OS (high: not reached vs intermediate: 300 days vs low: 375 days, p<0.001) in PD-1 inhibitor-treated patients.

In patients not treated with an anti-PD-1, TMB was not correlated with survival; thus, according to Dr. Johnson, TMB has a specific effect in immunotherapy-treated patients.

Jonathan E. Rosenberg, M.D., Memorial Sloan Kettering Cancer Center, reported on the results of an exploratory analysis of predictors of response in the IMvigor 210 Phase II trial12 that evaluated the efficacy of Tecentriq.

Patients (n=310) with metastatic urothelial carcinoma and who had progressed on prior platinum therapy were treated with Tecentriq (1200 mg IV Q3W until loss of benefit).

Dr. Rosenberg reported an improved efficacy among patients with higher PD-L1 expression (ORR: 28% versus 10%); however, durable responses to immunotherapy were seen regardless of PD-L1 status.

TMB was assessed using a small sequence of the genome (same 236-315 genes used by the previous two studies) as a proxy for the whole genome. Dr. Rosenberg reported that median mutational load was higher in patients who responded to Tecentriq (12.4 mutations/Mb) compared with non-responders (6.4 mutations/Mb, p<0.0001).

The patients within the highest quartile of median mutational load (>16 mutations/Mb for platinum treated and >13.5 mutations/Mb for first line cisplatin-ineligible) had higher OS (p=0.0012 and p=0.0079, respectively) than patients in the lower quartiles.

When included in a multivariable model, both PD-L1 and TMB were independent and significant predictors of response to Tecentriq (p=0.0109 and p<0.0001, respectively). 

The results of these three studies suggest that a high TMB is correlated with clinical response to immunotherapy. Each study used the same set of genes for analysis; however, each study provided a different threshold or definition of “high” TMB and reported that some patients with a lower TMB also have clinical responses to immunotherapy.

While assessing TMB has potential as a biomarker, a consensus will need to be made regarding which threshold of TMB should be used to identify those patients that would benefit most from immunotherapy treatment.

As well, the analysis by Dr. Rosenberg included both PD-L1 and TMB in addition to other tumor-specific characteristics in combination and reported that each factor was a significant and independent predictor of response to immunotherapy.

The result of that analysis suggest that simultaneous assessment of PD-L1 status and TMB as well as other tumor specific factors may be necessary to better define the patient population that will best benefit from immunotherapy.

Mismatch Repair Deficiency

Two abstracts reported updated information with more patients from the same study13 that focused on mismatch repair (MMR) deficiency or microsatellite instability (MSI) as markers of immunotherapy response.14, 15

The original data supported the FDA Breakthrough Therapy designation for Keytruda in relapsed/refractory metastatic colorectal cancer patients with MSI-high tumors 13.

MMR deficiency, due to defects commonly observed in MLH1, MSH2, MSH6, and PMS2, can lead to microsatellite instability that is present in either the germline (as in Lynch syndrome) or is sporadic in nature.  

In the first abstract,14 30 patients with MMR-deficient metastatic or locally advanced endometrial, ampullary or biliary, pancreatic, small bowel, gastric, prostate, sarcoma or thyroid cancers who had received at least one prior therapy were treated with Keytruda (10mg/kg every 14 days).

After a median follow-up of 10 months, 53% had an objective response, and responses were observed in all of the cancer types evaluated and with durable disease control. Unfortunately, patients with MMR-proficient cancers were not included in this analysis, thus this study is unable to determine the utility of testing for MMR as a biomarker for Keytruda.

The second abstract15 focused on metastatic colorectal cancer also treated with Keytruda (10mg/kg every 14 days), and included patients with both MMR-deficient (n=28) and MMR-proficient (n=25).

The MMR-deficient group had greater ORR (57% vs 0), PFS (not reached vs 2.3 months) and OS (not reached vs 5.98 months) than patients who were proficient in MMR.

In contrast to the previous response results for PD-L1 and TMB, the response to immunotherapy in MMR-proficient colorectal tumors was zero. Thus, MMR deficiency in colorectal cancer appears to be a selective biomarker for treatment with Keytruda.

More and larger studies will be necessary to further evaluate MMR deficiency as a possible biomarker of response to immunotherapy, as well as the relationship MMR deficiency as a biomarker has with PD-L1 expression and/or TMB.

Two trials investigating Keytruda treatment in MMR-deficient colorectal cancer include a Phase III trial of first-line therapy with Keytruda versus standard therapy in patients with MSI-high or MMR-deficient stage IV colorectal carcinoma (KEYNOTE-177, NCT02563002).

A pivotal Phase II trial is investigating Keytruda as a monotherapy in previously treated locally advanced unresectable or metastatic colorectal cancer with MMR deficiency or high microsatellite instability (KEYNOTE-164, NCT02460198).

Other immunotherapies in earlier phase trials that are currently recruiting include a Phase II study investigating treatment with Opdivo with or without Yervoy® (ipilimumab, Bristol-Myers Squibb) in recurrent and metastatic colorectal cancer with MSI-high tumors (CheckMate-142, NCT02060188) and a Phase I trial for durvalumab (MEDI4736, AstraZeneca/MedImmune; NCT02227667).

Interestingly enough, Roche will soon initiate a three-arm Phase III trial (NCT02788279), which will evaluate Tecentriq, Tecentriq plus the MEK inhibitor Cotellic® (cobimetinib, Roche/Genentech/Exelixis) or Stivarga® (regorafenib, Bayer/Amgen) in patients with colorectal cancer following two prior lines of cytotoxic chemotherapy for metastatic disease.

Although some patients will be MSI-high, a 5% cap for these patients was set. It will be interesting in light of the past data with Keytruda to see whether this trial will be successful, but if it is it would allow Tecentriq’s use in a less restricted patient population.

Conclusions

Among the FDA-approved indications for the three approved PD-1/-L1 inhibitors, only one currently limits use to an immunotherapy biomarker-defined patient subtype (Keytruda in PD-L1-overexpressing platinum-pretreated NSCLC).

Nevertheless, as the development of these agents is rapidly expanding across tumor types, there remains strong interest in identifying predictive biomarkers of response to these agents.

PD-L1 as a biomarker is dynamic rather than static, and patients treated with immunotherapies appear to respond to treatment regardless of expression status. MMR deficiency and tumor mutation burden as biomarkers also show evidence of differential response to PD-1/-L1 inhibitors.

Of the three possible biomarkers of PD-L1, TMB and MMR deficiency, MMR deficiency has the strongest evidence supporting its predictive behavior to PD-1 inhibitors in MSI-high colorectal cancer. PD-L1 status and TMB may best work as biomarkers when combined with each other and potentially other tumor specific factors.

Thus, the current oncology model of using a single biomarker to select patients for treatment may not be the best approach in the era of immunotherapy.  

About the Contributor

Kantar Health is a leading global healthcare advisory firm and trusted advisor to the world’s largest pharmaceutical, biotech, and medical device and diagnostic companies. It combines evidence-based research capabilities with deep scientific, therapeutic and clinical knowledge, commercial development know-how, and marketing expertise to help clients launch products and differentiate their brands in the marketplace.

Kantar Health provides both the research and consulting services you need to support your business decisions across the product lifecycle for cancer therapies. Our oncology expert status has been earned because of the proprietary material we publish on cancer epidemiology, trends in oncology market access, oncology pipeline potential and treatment trends that help our clients with their business issues.

If you would like us to act as catalysts for you, contact us at www.kantarhealth.com.

References

  1. Topalian SL, et al. N Engl J Med. 2012; 366:2443-2454.
  2. Meng X, et al. Cancer Treat Rev. 2015; 41:868-876.
  3. Taube JM, et al. Sci Transl Med. 2012; 4:127ra37.
  4. Madore J, et al. Pigment Cell Melanoma Res. 2015; 28:245-253.
  5. Tumeh PC, et al. Nature. 2014; 515:568-571.
  6. Bhaijee F, Anders RA. JAMA Oncol. 2016; 2:54-55.
  7. Bass P, et al. J Clin Oncol. 2016; 34(suppl); abstr 9015.
  8. Herbst RS, et al. Lancet. 2016; 9; 387(10027):1540-50.
  9. Gadgeel SM, et al. J Clin Oncol. 2016; 34(suppl); abstr 9016.
  10. Spigel DR, et al. J Clin Oncol. 2016; 34(suppl); abstr 9017.
  11. Johnson DB, et al. J Clin Oncol. 2016; 34(suppl); abstr 105.
  12. Rosenberg JE, et al. J Clin Oncol. 2016; 34(suppl); abstr 104.
  13. Le DT, et al. N Engl J Med. 2015; 372(26):2509-20.
  14. Diaz LA, et al. J Clin Oncol. 2016; 34(suppl); abstr 3003.
  15. Le DT, et al. J Clin Oncol. 2016; 34(suppl); abstr 103.

 

 

 

Post a Comment

Your email is never shared. Required fields are marked *