Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards

ASCO University
Apr 13, 2016 9:33 AM

Participant Instructions: Welcome to the Molecular Oncology Tumor Board Series! This educational initiative is a collaboration between the American Society of Clinical Oncology (ASCO), College of American Pathologists (CAP), and Association for Molecular Pathology (AMP).

A new case will be presented each month with discussions led by an expert pathologist and medical oncologist. Have an interesting case in mind? Submit your hypothetical patient cases for consideration in an upcoming Molecular Oncology Tumor Board discussion forum.

This month’s topic is led by Drs. Leora Horn (Vanderbilt) and Mari Mino-Kenudson (Massachusetts General Hospital).

Users are encouraged to leave comments and post questions about the case in order to generate a wide discussion among the cancer care community. You can also receive email notifications when new comments are posted by clicking the “Follow this Conversation” option located at the bottom of this page.

When posting, please abide by the terms and conditions of this website.

Please see attached for supplemental resources related to the case.

Comments

12456

ASCO University
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 13, 2016 9:39 AM

Patient Case #1

Age/Sex: 72 year old, Caucasian, male former smoker

Medical History: Hypertension, diabetes, BPH

Type of Tumor: NSCLC adenocarcinoma histology, positive for the KRAS mutation G12C

Relevant Markers:  NGS shows KRAS mutation G12C

Prior Treatment History/Response: Treated with four cycles of carboplatin and pemetrexed followed by maintenance pemetrexed. After 16 months on maintenance pemetrexed has disease progression.

Images/Scans/Pathology: KRAS Mutant HE

12461

ASCO University
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 13, 2016 9:41 AM

Discussion Questions

  1. What are available second line therapy options?
  2. Do you test for PD-L1 prior to recommending therapy?
  3. Does smoking history affect your treatment recommendations?

12466

Benjamin Philip Levy, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 13, 2016 11:39 AM

For me, outside a clinical trial, this patient warrants a checkpoint inhibitor with a PD1 drug given that there is tumor progression on chemotherapy. Clearly we now have data from Checkmate 057 that nivolumab offers meaningful improvement in survival vs. docetaxel. This benefit did seem to be restricted to patients whose tumors' expressed PDL1 by IHC. That said, whether PDL1 testing is the right enrichment strategy for this patient population and whether PDL1 negative patients should be excluded from treatment with these agents are outstanding questions.  At our institution we are currently NOT testing for PDL1 by IHC routinely, but I would be curious to others' practice in both academic and community settings. I would offer this patient Nivolumab but if was tested for PDL1 and positive, then perhaps pembrolizumab. Would also love to hear thoughts on KRAS directed therapies in development outside of selumitinib.  

12471

Alexander E. Drilon, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 13, 2016 3:03 PM

There are multiple second-line options for this patient that include treatment regimens such as docetaxel and ramucirumab. Given the patient's smoking status and the presence of a KRAS transversion mutation that is a common molecular signature in tumors from smokers, I agree with Dr. Levy. My preferred approach would be to administer an immune checkpoint inhibitor. I do not routinely test for PD-L1 in the clinic due to a variety of factors: PD-L1 status is dynamic, the definition of ‘positivity’ varies from assay to assay, and most importantly, responses to single-agent immune checkpoint inhibition are not exclusively limited to PD-L1-positive tumors. Several KRAS G12C-specific covalent inhibitors are currently in early-phase laboratory testing and, looking to the future, the hope is that these will emerge as viable treatment options for patients once they hit the clinic.

12476

Nathan A. Pennell, MD, PhD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 13, 2016 3:14 PM

I agree that the PD-1 inhibitor nivolumab would be my next choice based on the Checkmate 057 trial, and I also would not check for PD-L1 ahead of time because it is not clear that the biomarker is meaningful in choosing the next line of therapy. Yes, higher PD-L1 staining increased the magnitude of the superiority over docetaxel (the alternative choice), but even PD-L1 negative tumors had a comparable response rate to docetaxel (9%) and there was no significant difference in OS between the two. As docetaxel is quite toxic, I think simply treating second line patients with nivolumab makes the most sense and reserve chemo for third line. If the PD-L1 test were done and strongly positive then pembrolizumab would be reasonable, but since testing is not routine in many places this doesn't come up much.

If the patient was a nonsmoker, though, I would think twice since these patients seem to have a lesser chance of response to PD-1 therapies. It is still not clear that they would be better off with chemo but would be worth more of a discussion.

12481

Anis Toumeh, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 13, 2016 6:27 PM

I am mostly interested in learning about what other practices (academic / community) are doing regarding PD-L1 testing. I don't do it routinely, and even in CheckMate 057, although Nivolumab did not seem to perform better than Docetaxel survival wise in the PD-L1 negative cohort, it was much more toleratble. I also agree that PD-L1 seems to be a dynamic marker. 

12486

Leora Horn, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 14, 2016 8:36 AM

Course Faculty Response

I don’t think PD-L1 is so much a dynamic biomarker as much as a poor biomarker. Keynote 10 showed responses to pembrolizumab both in fresh and archival specimens. We have become accustomed to using biomarkers in NSCLC, perhaps more so than any other disease. A biomarker that has a response rate of about 30% in a positive cohort and a response rate of 10% in a negative cohort (similar to chemotherapy) is not a good decision tool. Especially when you think about the response to TKIs seen in patients with tumors that are positive for EGFR and ALK compared to patients with tumors that are wild type. 

12491

Christopher G. Azzoli, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 15, 2016 1:36 PM

There is no drug for KRAS, and yet no one would argue with testing this patient's tumor for KRAS because it better classifies his lung cancer for drug selection, prioritization, and opens doors for clinical trial participation.  PDL1 IHC by 22C3 has practical value.  If positive, the patient would be more likely to benefit from pembrolizumab, would have access to 2 immune checkpoint inhibitors instead of just nivolumab, and it might impact clinical trial participation.  The only reason I can think of not to check PDL1 IHC in our patients is cost.  We should not short-change our patients if it gets them more choice and opportunity.

12496

Nathan A. Pennell, MD, PhD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 15, 2016 4:19 PM

My primary concern about PD-L1 testing isn't that it might lead someone to choose one PD-1 therapy over another, but rather that a NEGATIVE result might lead oncologists to recommend against PD-1 directed therapies entirely. As Dr. Horn pointed out, 2/3 of high PD-L1 expressors will not respond to checkpoint inhibitors, while 10% of patients with tumors lacking PD-L1 expression will respond, which is comparable to chemotherapy but with less toxicity (and more durable control in responders).

In a perfect world yes I would like to have that information (PD-L1 expression) on every patient but I fear that widespead testing now will end up inappropriately suggesting that this marker is valid for choosing the best treatment, and that positive pts will get PD-1 treatment and negative ones will get chemo, when that is not what the data suggests is the best strategy. Until we have first line data in PD-L1 positive patients where we really do need the marker, I would skip it and save the expense.

12506

Mari Mino-Kenudson, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 18, 2016 9:13 AM

Course Faculty Response

1. What are the available second line therapy options?

Please refer to Dr. Horn’s response.

2. Do you test for PD-L1 prior to recommending therapy?

There are multiple antibody clones, immunohistochemistry (IHC) platforms and scoring systems for PD-L1 expression, and each anit-PD-1/PD-L1 agent is coupled with a specific antibody clone, IHC platform and scoring in the clinical trials.  Of those, membranous PD-L1 expression of any intensity on > 50% of the tumor cells (Figure - KRAS mutant PD-L1 score 2) confirmed by clone 22c3 with Dako automation is now required for treatment with pembrolizumab. Although the biomarker assessment of Checkmate 057 (nonsquamous NSCLC) has shown that PD-L1 expression on at least 1% of tumor cells (Figure – PD-L1 score 1x200) confirmed by clone 22-8 with Dako automation were associated with higher response rates to nivolmab and overall survival when treated with nivolmab compared to docetaxel, PD-L1 testing is not required for treatment of NSCLC patients with nivolmab. Of note, there was no difference in response to nivolmab between PD-L1 positive and negative groups in squamous NSCLC patients (checkmate 017). If treatment with pembrolizumab is considered, the patient’s tumor tissue needs to be examined with the specific IHC assay (clone 22c3 with Dako automation), but the issues is availability of the IHC platform since it is difficult for pathology laboratories to equip multiple, different IHC autotimers.  If the Dako autostainer is available in the pathology laboratory, IHC with clone 22c3 will likely be performed on site, but if the IHC platform is not available, the patient’s tumor sample will likely be sent out to a commercial laboratory that offers the specific PD-L1 IHC assay.

3.    Does smoking history affect your treatment recommendations?

Significant smoking history is reportedly associated with a large mutation burden in the tumor that has been liked to immune microenvironment of the tumor. It has been shown that cytotoxic T lymphocytes (CTLs) infiltrating the tumor cells (Figure - KRAS Mutant CD8) secrete interferon  that subsequently induces PD-L1 expression on the tumor cells, immune cells and others as well as PD-1 on T cells resulting in acquired immune resistance. Thus, PD-1/PD-L1 axis blockade is more likely to be effective in tumors from smokers (associated with the tumor immune microenvironment) than in those with never or light smoking history.  

12511

Leora Horn, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 18, 2016 9:17 AM

Course Faculty Response

1. What are the available second line therapy options?

There are now 5 FDA second line FDA approved agents for patients with NSCLC pemetrexed (which this patient already had), docetaxel +/- ramucirumab, erlotinib, nivolumab, and pembrolizumab in patients with tumors that are PD-L1 positive.

Two large randomized phase III trials compared docetaxel to immune checkpoint inhibitors in patients with NSCLC, adenocarcinoma histology who had progressed following platinum based chemotherapy.

2. Do you test for PD-L1 prior to recommending therapy?

Checkmate 057 compared docetaxel 75 mg/m2 every 3 weeks to nivolumab 2 mg/kg every 2 weeks in nonsquamous NSCLC patients who had progressed on platinum-based chemotherapy, EGFR and ALK positive patients must also have progressed on a TKI. This trial met its primary endpoint of improvement in overall survival for patients treated with nivolumab compared to docetaxel (median OS 12.4 vs. 9.4 months). In addition the 12 month and 18 month survival was superior for nivolumab 51% and 39% vs. 39% and 23% respectively. This trial enrolled patients regardless of PD-L1 expression. The overall responses were higher for nivolumab treated patients 20% vs 9% for docetaxel. Responses were higher in patients who were PD-L1 positive regardless of level of PD-L1 expression; 1%, ≥5% and ≥10%. Overall survival was also higher in patients who were PD-L1 positive treated with nivolumab compared to docetaxel. While survival was similar in patients who were PD-L1 negative.

Keynote010 compared docetaxel 75 mg/m2 every 3 weeks to pembrolizumab 2 mg/kg or 10 mg/kg every 3 weeks in both squamous and nonsquamous NSCLC patients who had progressed on platinum-based chemotherapy, EGFR and ALK positive patients must also have progressed on a TKI. Patients enrolled had to be positive for PD-L1 expression. This trial met its primary endpoint of improved overall survival for patients treated with pembrolizumab 10.4 months with 2 mg/kg, 12.7 months with 10 mg/kg compared to 8.5 months for docetaxel. Among patients with tumors that were ≥ 50% PD-L1 positive overall survival was even longer 17.3 and 14.9 months at 10 mg/kg and 2 mg/kg respectively compared to 8.2 months for docetaxel. The overall response rates were 18% for pembrolizumab at both doses versus 9% for docetaxel. In patients with tumors that were ≥ 50% PD-L1 positive responses were 30% and 29% in pembrolizumab treated patients versus 9% for docetaxel.

In lung cancer patients where tissue is not always available, testing is possible but not required for therapy with nivolumab. While testing is required for treatment with pembrolizumab.

3. Does smoking history affect your treatment recommendations?

Prior data has suggested that immunecheckpoint inhibitors have greater efficacy in smokers compared to never smokers. In checkmate 057 the HR for overall survival was 0.70 (0.56-0.86) in current and former smokers compared to 1.02 (0.64-1.61) in never smokers. Prior data presented by Dr. Matthew Helllman with pembolizumab from the Keynote 001 trial showed similar results.

12516

ASCO University
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 18, 2016 9:22 AM

Patient Case #2

What if the patient is different:

  1. Patient is a lifetime never smoker with adenocarcinoma histology and no identifiable driver mutations
  2. Patient is a lifetime never smoker with adenocarcinoma histology positive for the EGFR mutation deletion 19 and rather than progression on first line chemotherapy has progressed on first line erlotinib

12521

ASCO University
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 18, 2016 9:23 AM

Discussion Questions

  1. What second line therapy do you offer for a never smoker?
  2. What do you do for a patient who has progressed on a first line EGFR TKI is immune checkpoint inhibitor appropriate?

12526

Anis Toumeh, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 19, 2016 8:43 PM

Subgroup analysis from CheckMate 057 and Keynote 010 questioned the benefit of immunotherapy in never smokers and in those positive for EGFR mutations. One could consider testing for PD-L1 in these circumstances to determine who is likley to benefit from immunotherapy. I am interested in hearing what others think about re-biopsy and testing for PD-L1 in those patients, keeping in mind the issues of the marker itself mentioned earlier. 

12531

Nathan A. Pennell, MD, PhD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 20, 2016 8:29 AM

That is a great question, I am not aware of whether we know if the lesser chance of response to PD-1 inhibitors in nonsmokers is (at least partly) related to lower expression of PD-L1, or if the lower response rate is independent of PD-L1 expression. Another way of rephrasing this question is whether PD-L1-positive nonsmokers have the same or lower chance of response as a PD-L1-positive smoker? Does anyone have the answer to that? I am not sure that data was part of the Checkmate 057 or Keynote 010 presentations.

12536

Leora Horn, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 20, 2016 9:48 AM

Course Faculty Response

There was data from Keynote OO1 indicating that if you are a never smoker and PD-L1 positive the response to pembrolizumab was higher with a durable progression free survival compared to patients who are never smokers and PD-L1 negative.

12551

Mari Mino-Kenudson, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 22, 2016 9:49 AM

Course Faculty Response

1. What second line therapy do you offer for a never smoker?

Please refer to Dr. Horn’s response.

2. What do you do for a patient who has progressed on a first line EGFR TKI is immune checkpoint inhibitor appropriate?

The mechanisms for resistance to EGFR TKI have been well studied. Approximately half of patients with EGFR-mutated NSCLC who developed resistance to a first line EGFR TKI have an acquired T790M EGFR mutation. The other mechanisms include MET amplifications (5-15%) and histologic transformation to small cell carcinoma (7%). Patients with those resistant mechanisms could be treated with targeted agents either commercially available or in the trial setting, or treatment for small cell carcinoma. If no resistant mechanism is identified by repeat biopsy and the patient’s tumor shows PD-L1 expression, treatment with an immune checkpoint inhibitor is conceivable, but its response may not be significant. The subset analyses of Checkmate 057 and Keynote 010 have shown that patients with EGFR mutation positive NSCLC experienced shorter overall survival compared to those with EGFR wild-type tumor when treated with nivolumab and pembrolizumab, respectively. 

12556

Leora Horn, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 22, 2016 9:51 AM

Course Faculty Response

In never smokers and patients who are EGFR and ALK positive the data with immune checkpoint inhibitors has been less promising. A subset analysis as mentioned previously from checkmate 057 demonstrated a HR of 1.02 for never smokers treated with nivolumab compared to docetaxel. Response rates were also lower for never smokers treated with nivolumab compared to responses with docetaxel (10% vs. 16%). Similarly for patients with EGFR mutation positive NSCLC the HR was 1.18 (0.69-2.0) compared to 0.66 (0.51-0.86) in patients who were EGFR wild type. Similarly in Keynote010 the HR was 0.88 (0.45-1.70) for patients with tumors that were EGFR mutation positive compared to 0.66 (0.55-0.80) for patients with tumors that were EGFR wild type.

Therefore for never smokers who are EGFR mutation positive and have progressed on an EGFR TKI it is important to obtain a biopsy and if positive for the T790M mutation, there is currently one FDA approved targeted therapy, osimertinib and many in clinical trial. Similarly for patients who are ALK positive and have progressed on crizotinib, alectinib and ceritinib are FDA approved second line options. If patients are T790M negative or do not have an actionable mutation it would be reasonable to test for PD-L1 expression and if positive consider treatment with an immune checkpoint inhibitor but if negative chemotherapy would be preferred.

12581

Mari Mino-Kenudson, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 27, 2016 9:49 AM

Course Faculty Summary

  • To date, at least two different mechanisms of PD-L1 expression on tumor cells have been described: innate immune resistance and adaptive immune resistance. In the former, the up-regulation of PD-L1 expression is secondary to constitutive oncogenic signaling within tumor cells including EGFR mutations and ALK rearrangements. By contrast, in adaptive immune resistance, PD-L1 expression is induced on tumor cells secondary to local inflammatory signals, which in turn induces PD-1 on T cells. When engaged by PD-L1 or other ligands, PD-1 inhibits kinases that are involved in T cell activation through the phosphatase SHP250 leading to apoptosis of T cells, although additional signaling pathways are likely also induced. Thus, blockade of PD-1/PD-L1 interaction will reinstitute the active anti-tumor immune response. KRAS-mutated NSCLC, especially those with significant smoking history, may harbor large mutation burdens (large numbers of neoantigens) leading to adaptive immune resistance (due to immune microenvironment with significant local inflammatory signals).  Growing evidence suggests that KRAS-mutated NSCLC are more likely to have PD-L1 expression and respond to the PD-1/PD-L1 blockade than KRAS wild type.
  • There are currently two PD-L1 IHC assays approved by FDA as either companion or complementary diagnostics, and two more (those used with atezolizumab and durvalumab) are expected to be approved by FDA in 2016. The presence of multiple IHC assays and the fact that each assay is coupled with a specific anti-PD-L1/PD-1 agent has brought complexity to PD-L1 testing. Two projects (Blue print and NCCN Bristol-Myers Squibb PD-L1 Project) are currently comparing the four PD-L1 IHC assays +/- a commercial antibody (clone E1L3N) with respect to targets, intensities, and frequency of staining, among others, and they may help understanding differences in the performance of those assays, and possibly establishing the standard for PD-L1 testing. 
  • Several mechanisms of resistance to 1st line EGFR TKIs have been identified and the majority of those are actionable. Thus, in patients with EGFR-mutated NSCLC who progressed on an EGFR TKI, repeat biopsy should be performed to identify a resistant mechanism.

 

12586

Leora Horn, MD
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 27, 2016 9:51 AM

Course Faculty Summary

  • Immune checkpoint inhibitors are approved as second line therapy in patients with both squamous and nonsquamous NSCLC.
  • In patients with nonsquamous NSCLC PD-L1 expression has been associated with improved response rates and overall survival. However in patients with tumors that are negative for PD-L1 expression responses and survival with nivolumab are similar to docetaxel with reduced toxicity.
  • There are currently two different FDA approved assays for assessment of PD-L1 expression with different anti-bodies and different cut-offs. The IASLC in conjunction with the FDA is comparing the different assays that have been evaluated in clinical trials including those used with atezolizumab and durvalumab which may help sort out the appropriate assay and cutpoint to use.
  • In patients who are EGFR mutation positive and progress on an EGFR TKI a biopsy should be performed and if positive for T790M a targeted therapy, osimertinib should be offered.
  • In never smokers or patients who are EGFR or ALK positive who do not have a targeted therapy option PD-L1 testing should be considered prior to starting an immune-checkpoint inhibitor.

 

12591

ASCO University
Re: Non-small Cell Lung Cancer (PD-L1 Expression): Molecular Oncology Tumor Boards
Apr 27, 2016 9:54 AM

Thank you to Drs. Horn and Mino-Kenudson for leading the discussion of this case and also to all of those who contributed to the conversation! The forum is now closed to further comments but participants can claim credit through ASCO University.

Please check back in mid-May for a new case in this series related to gastric cancer.    

Have an interesting case in mind? Submit your hypothetical patient cases for consideration in an upcoming Molecular Oncology Tumor Board discussion forum. 

 


Advertisement