Driving Progress in HNSCC: Integrating Immunotherapy and Novel Targeted Therapies Across the Disease Continuum

 

Let us turn to recurrent metastatic disease. This is a man I care for. Three years ago he came in with a T4N2 p16+ oropharynx cancer, and we treated him with conventional chemoradiation. He achieved a complete response, but now he presents with bilateral lung nodules, the largest of which is 1.8 centimeters in size.

 

We use the Naveris test to measure TTMV-HPV, and this was 38. We undertook a CT-guided lung biopsy, which was positive for a p16+ squamous cell carcinoma. His PD-L1 level was 9.

 

[00:22:24]

 

Pretest 2

 

In your current practice, would your recommendation for this patient's initial treatment be:

 

1. Cetuximab with platinum and 5-FU;
2. Pembrolizumab monotherapy;
3. Pembrolizumab together with chemotherapy consisting of platinum and 5-FU;
4. A doublet of 2 immune-agents, nivolumab and ipilimumab; or
5. The PD-1 inhibitor, dostarlimab.

 

It looks like 82% of you would give pembrolizumab with platinum and 5-FU.

 

[00:23:10]

 

Systemic Therapy Recommendations for Non-nasopharyngeal HNSCC

 

These are in fact the NCCN guidelines, which say that for recurrent unresectable or metastatic head and neck squamous cell carcinoma, not amenable to curative surgery or radiation. First-line therapy should be pembrolizumab/platinum and 5-FU, where the platinum can be carboplatin or cisplatin. Pembrolizumab is also an option for tumors that express PD-L1 with a CPS 1 or higher.

 

In subsequent lines of therapy, if there has been no prior immunotherapy, the patient should receive nivolumab or pembrolizumab. Other options are also provided though, and those include combination regimens such as:

 

• Cetuximab/platinum and 5-FU, which I think most of us reserve for those patients who are PD-L1 negative;
• Cisplatin/cetuximab;
• A platinum and a taxane;
• Cisplatin with 5-FU;
• Platinum/taxane/cetuximab; or
• Pembrolizumab/platinum and taxane.

 

We will talk a little bit more about that regimen as well.

 

Then for patients who are not candidates for multi-drug chemotherapy, there are a number of single agents that have activity. In certain circumstances like cutaneous cancers, sinus cancers, there are other regimens that are looked at. As well for patients with FGFR mutations, you can think about erdafitinib. For patients that are HER2-expressing, T-DXd.

 

[00:24:40]

 

KEYNOTE-012: Overall Response by PD-L1 Status

 

Let us go back to the evidence for PD-1 inhibition. We will start with KEYNOTE-012, which was really the first large study to look at head and neck squamous cell carcinoma patients with pembrolizumab. This was a phase Ib trial. Patients were enrolled actually irrespective of PD-L1 expression. If you look at the PD-L1 positive patients, they had a response rate of 18% and the PD-L1 patients had a response rate of 19%, but that was if you were only assessing PD-L1 expression on tumor cells.

 

We now know that it is also informative to look at PD-L1 expression on immune cells that generates a much larger population of patients who are PD-L1 positive, but the ones who are PD-L1 negative on this assay are really a different population. In KEYNOTE-012, we saw only a 4% response rate for pembrolizumab in those patients.

 

[00:25:35]

 

KEYNOTE-012: Phase Ib Trial of Pembrolizumab in Recurrent or Metastatic HNSCC

 

This shows you that we had responses both for HPV-positive patients shown as purple lines, and HPV-negative patients shown as green lines. The majority of the responses were durable and many of them were quite deep.

 

There were a few patients with what appeared to be rapid progression and most of those patients had HPV-negative disease.

 

[00:25:59]

 

CheckMate 141: Nivolumab in Recurrent/Metastatic HNSCC After Platinum Therapy

 

There was another study also done in platinum-treated patients, the CheckMate 141 trial. Here, patients got nivolumab after platinum therapy. They were stratified by prior cetuximab. If they were not randomized to nivolumab, the investigator had a number of choices, methotrexate, docetaxel, cetuximab with a primary endpoint of overall survival.

 

[00:26:22]

 

CheckMate 141: OS for Nivolumab vs Investigator’s Choice in Recurrent/Metastatic HNSCC

 

You see here that this was a positive trial. The median overall survival increased from 5.1 to 7.7 months. The objective response rate from 6% to 13.3%. Again, it was notable how long lasting some of these responses were with still a substantial difference at 18 months.

 

[00:26:42]

 

KEYNOTE-048: Pembrolizumab ± Chemotherapy vs EXTREME in R/M HNSCC

 

The success of PD-1 inhibition in the previously treated patients made us curious to see whether or not this could have an impact earlier in the course. KEYNOTE-048 was a trial that aimed to establish the role of PD-1 inhibition in the first-line setting.

 

We understood that we were comparing a regimen with a response rate of about 18% in pembrolizumab to regimens with 35% response rates in the control arm of chemo with cetuximab.

 

We took a 2-pronged approach to understanding where the liabilities there might be. The first was recognizing that response was higher as the CPS PD-L1 expression went higher. We started with an analysis where the patients with the highest biomarker expression were analyzed first, then all those PD-L1 positive patients and then all patients. As well we included an arm where pembrolizumab was given together with chemotherapy assuming that the response rate there would be higher.

 

[00:27:46]

 

KEYNOTE-048: OS for Pembrolizumab Alone vs Cetuximab/Chemotherapy

 

In fact, we did see a benefit for the use of pembrolizumab in the first-line setting for recurrent metastatic head and neck cancer. If you looked at the entire population, the hazard ratio was 0.81 and a 48-month survival went from 6.6% to 15.4%. If you looked at all PD-L1 expressing patients, so that is on the left there.

 

CPS 1 patients, median survival went from 10.4 to 12.3 months. The hazard ratio there was 0.74. The P value was .0008. You can see that 4-month survival went from 6% to 16.7%.

 

Then as we predicted in the CPS 20 population, the benefit was somewhat greater. OS went from 10.8 to 14.9 months. Hazard ratio, 0.61, and 4-year survival going from 8% to 22%.

 

[00:28:46]

 

KEYNOTE-048: OS for Pembrolizumab/Chemotherapy vs Cetuximab/Chemotherapy 

 

If we look now at the pembrolizumab plus chemotherapy group, here again there looked like there was a benefit for the entire population. Median survival went from 10.7 to 13 months. Hazard ratio, 0.71. In the CPS 1 over here, it went from 10.6 to 13.6 months. Four-year survival going from 4% to 22%. Here the hazard ratio was 0.64.

 

For the CPS 20, the highest expressing median survival went from 11.1 to 14.7 months, and 4-year survival at 28.6% there.

 

[00:29:28]

 

KEYNOTE-048: OS in Subgroups for Pembrolizumab vs EXTREME

 

If we looked across subgroups, there was no group where it seemed as if pembrolizumab was harmful. The benefit was seen in HPV positive and negative disease, and it was seen for those patients who had metastatic or recurrent disease.

 

[00:29:49]

 

KEYNOTE-048: OS by Subgroups for Pembrolizumab + Chemotherapy vs EXTREME

 

If we look in the pembrolizumab plus chemotherapy group here, again, there is no group that was deleterious. This time there is also no group where there was not a benefit. Here you can see particularly striking benefits in the CPS 20 group for the HPV-positive metastatic patients.

 

[00:30:12]

 

KEYNOTE-048: Long-term OS Results

 

We now have the 5-year result. Five-year results were just reported in The European Journal of Cancer by Makoto Tahara. You can see that the benefit continues even at 5 years with a substantial improvement in overall survival for these patients with recurrent metastatic head and neck squamous cell cancer.

 

[00:30:32]

 

CheckMate-651: Nivolumab + Ipilimumab vs EXTREME as First-line Therapy for R/M HNSCC

 

It is always awkward to present data when the leader of the trial is sitting right next to you. But this is Robert's trial, CheckMate-651, looking at nivolumab plus ipilimumab vs the EXTREME regimen as first-line therapy. This built on the notion that PD-1 inhibition was beneficial and added a second immunotherapy agent, the second immune checkpoint inhibitor, ipilimumab, which targets CTLA-4 and is thought to be priming, I think with the hypothesis that this might overcome the lower benefit in the PD-L1 negative patients.

 

Here, patients were randomized 1-to-1 to the nivolumab-ipilimumab combination or to the same control arm used in CheckMate-048.

 

[00:31:18]

 

CheckMate-651: OS With Nivolumab + Ipilimumab vs Cetuximab-Based CT (SoC) in R/M HNSCC

 

As you can see here in all randomized patients, first of all the standard of care was doing better in this population. I am not sure that everybody comes in their mind to the same explanation for this. I think that in the wake of KEYNOTE-048, patients who really needed chemotherapy for a higher response rate were less likely to be enrolled, but we cannot be certain. But here we saw a standard of care median survival of 13.5 going to 13.9 for the doublet.

 

If we confined the analysis to the PD-L1 CPS 20 population, the standard of care had 14.6 months and this went to 17.6. But of course this is a smaller group, and it was not powered prospectively for this analysis. So that P value of .04 does not quite make statistical significance. Although you can see looking at the curves, you would probably rather be on that top curve.

 

[00:32:12]

 

CheckMate-651: PFS, DoR With Nivolumab + Ipilimumab vs SoC (All Randomized Patient)

 

Duration of response and progression-free survival are shown here. As we had seen in KEYNOTE-048, the progression-free survival for immunotherapy alone is a little bit shorter, but those patients who have a response to immunotherapy have very remarkable durations of response. Here for all comers it was 16.6 months.

 

[00:32:34]

 

LEAP-010: First-line Pembrolizumab ± Lenvatinib for R/M HNSCC

 

There was also rationale to think that if we could combine immunotherapy with anti-angiogenic, that this would have a beneficial effect both on ingress of T-cells as well as on removing some signals that foster suppressive cells in the microenvironment.

 

A randomized trial was undertaken to look at the addition of the anti-angiogenic TKI lenvatinib to pembrolizumab. This was a phase III randomized trial, a primary endpoint PFS, and as well as a co-primary of objective response rate.

 

[00:33:12]

 

LEAP-010: Response

 

Here the addition of lenvatinib to pembrolizumab raised, and I think we had a similar thing going on with the patient population here. You can see how well the control group does. In addition to the fact that patients knew they might get pembrolizumab monotherapy on this study, there were also a number of rules to exclude patients with a bleeding risk because of the use of lenvatinib.

 

Here, the control arm had a response rate of 25% including 9.8% CRs. This increased to 46% with the addition of lenvatinib. The CR rate also went up. Duration of response was quite nice.

 

[00:33:47]

 

LEAP-010: Overall Survival at IA2

 

However, overall survival indicated that there was crossover once you stopped the anti-angiogenic. The median survival of 17.9 months in the pembrolizumab monotherapy arm, it is obviously much better than we had seen in either of the prior trials and 15 months with the addition of lenvatinib. This regimen did not move forward.

 

[00:34:11]

 

Alternate Approaches

 

We also have some alternative approaches. PD-1 inhibition can be given in combination with alternate chemotherapy. The reference chemo regimen that has been around for a long time of platinum and infusional 5-FU that comes from Yanver Morgan's[?] EXTREME study is not popular either with patients or prescribers. It involves a port, 4 days of infusional 5-FU, a relatively higher hospitalization rate.

 

There has been a trial called the B10 study, single-arm study but relatively well sized, looking at whether or not the alternate chemo combination of platinum and paclitaxel could be substituted for the platinum and 5-FU. This appears to be quite feasible. The response rate in this trial was 43%, median overall survival 12 months. This is rather widely used nowadays.

 

[00:35:04]

 

KEYNOTE-B10: First-line Pembrolizumab + Carboplatin + Paclitaxel for R/M HNSCC

 

This shows you the objective response rate by demographic criteria here. You can see the higher response rates seen in patients with metastatic disease, HPV positive disease and better performance status. As I mentioned, PFS 5.6, overall survival 13.1 months, and duration of response as in all of the chemo combinations, a little bit shorter than for pembrolizumab monotherapy at 5.5 months.

 

I think it is an open question whether this is because chemotherapy is slightly immunosuppressive in the long run, but it is important to bear in mind that the denominator for these analyses of duration of response in the pembrolizumab chemotherapy arms is more than twice as high because there are some patients who are responding to chemo and not to immunotherapy.

 

[00:36:02]

 

FRAIL-IMMUNE: First-line Durvalumab + Carboplatin + Paclitaxel for R/M HNSCC in Patients Ineligible for Cisplatin

 

We also saw from France, the FRAIL-IMMUNE using the PD-L1 inhibitor durvalumab together with combination chemotherapy. Here, median overall survival of 18 months, 12-month survival of 63%, 2-year survival of 45% and an objective response rate of 71%. This was a group of patients with more elderly, more patients with comorbidities. This is a regimen I think that informs us particularly about the use of carboplatin and paclitaxel, given that durvalumab is not approved in this setting in head and neck cancer.

 

[00:36:41]

 

KEYNOTE-012 Phase Ib Trial of Pembrolizumab in R/M HNSCC: Response Biomarkers and HPV Status

 

Again, it is a pleasure to be presenting Robert's data in an analysis from KEYNOTE-012 trying to get at some of the biomarkers that can indicate who is going to benefit from PD-1 inhibition beyond just PD-L1 expression. Here you can see that across p16 status, tumor mutation burden, there was an indication that higher TMB, and inflamed gene expression profile were indicative of a higher likelihood of response to pembrolizumab.

 

[00:37:17]

 

Mutational Load: Genetic Alterations in HNSCC

 

Mutational load has also been looked at very extensively in the durvalumab trials. You see below there a randomized comparison of chemotherapy to durvalumab or durvalumab plus a CTLA-4 drug. In overall analysis, actually standard of care chemotherapy won out in this analysis. But if you look at the separation of the curves in the patients who have low tumor mutation burden, they sort of run together. But if you look over here at the patients who have higher tumor mutation burden, which for the purposes of this study was defined as 16 mutations per megabase. I think FDA approvals have been using 10.

 

But you can see that there is an advantage for the durvalumab arm here over the durvalumab/tremelimumab arm.

 

Again, another study that does not give us a clear signal about what to do about the CTLA-4 inhibitors.

 

[00:38:13]

 

KEYNOTE-048: OS Estimates Pembrolizumab Alone vs Cetuximab/Chemotherapy (CPS <1 and CPS 1-19)

 

We wanted to go back and learn more about PD-L1 expression. The statistical design of KEYNOTE-048 really focused on moving that cut point down from CPS 20 to CPS 1 to all patients and did not provide a lot of power for the analysis in the CPS less than 1 or the CPS 1 to 19 groups. But it was important to understand what we ought to do for those.

 

If we start on the left-hand side, this is CPS less than 1, and at the top is pembrolizumab alone. You can see here that pembrolizumab underperforms chemotherapy and cetuximab with a median overall survival of 7.9 compared to 11.3 months and a hazard ratio of 1.51.

 

Although this study was not powered for that analysis, I do not think these curves make you feel very good about giving pembrolizumab monotherapy in the first-line.

 

If you look at the 1 to 19 population, though the story is different. The overall survival goes from 10.1 to 10.8 months. The hazard ratio is 0.86. You can see that the pembrolizumab arm is somewhat at least numerically outperforming the control arm, even out here at the 3-year mark.

 

Turning to pembrolizumab and chemotherapy, this did look better than chemotherapy and cetuximab across all the groups. 10.7 to 11.3 in the CPS less than 1, and 9.9 to 12.7 in the 1 to 19. Even though we have less power for these analyses, I hope you find them informative.

 

[00:40:00]

 

CheckMate-651: OS by p16 and PD-L1 Status

 

There is also a question about the natural impulse for an HPV negative patient who is relatively low volume metastatic disease and a great performance status is to want to not give them chemotherapy. Clearly, we have some evidence that pembrolizumab benefits them, but the benefit actually seems greatest for pembrolizumab plus chemo in this group.

 

I wanted in this context to also pull out the p16 PD-L1 status data from CheckMate-651, where you see for the oropharynx cancer patients who are p16 positive, although the error bars cross 1 here, you do have a suggestion from the hazard ratio that they were benefiting from chemotherapy maybe a little bit more than from ipilimumab-nivolumab. If you think back to KEYNOTE-048, where we have that remarkable hazard ratio of 0.4 for chemo pembrolizumab in the CPS 20 patients who are HPV-positive, I think that is a reminder of how sensitive HPV-positive cancers are to DNA damaging agents.

 

Then in CheckMate-651 across the groups, if there was PD-L1 expression, it favored treatment with immunotherapy.

 

[00:41:16]

 

Select Novel Targeted Therapy Approaches for R/M HNSCC

 

There are other things that we are exploring. We would like to get past OSS of even the best case from LEAP-010 17 months. People are looking at combinations with tyrosine kinase inhibitors, the multi targeted kinase inhibitors, zanzalintinib is in a phase III trial. For HPV-negative patients and perhaps particularly those who are treatment resistant, there is overexpression of the receptor tyrosine kinase, c-MET.

 

You see here a randomized trial of ficlatuzumab that targets the MET ligand HGF. There are a large number of bispecifics now that attempt to go back to cetuximab, which after all was a good therapy for many years and performed not that much worse than ipilimumab-nivolumab in the CheckMate-651.

 

Look at co-targeting it with specific mechanisms of resistance. We have petosemtamab, which is a bispecific to EEG and LGR5 targeting wind signaling. Ficerafusp compound, co-targeting EGFR and TGF-beta, well known to be associated with treatment resistance to HER family targeting.

 

Then coming from the lung world, amivantimab, a bispecific for EGFR and MET. All of these are now moving into phase III trials. As well, we are learning from our colleagues in the breast and bladder and lung worlds that even if you have an antibody that does not have a very high response rate, it is very good at homing to the antigen on cancer cells. That this can be used to deliver chemotherapy in constructs that are called antibody drug conjugates.

 

We have some here that are targeted to B7-H3, EGFR and MET, and seeing responses with all of these. There are others, HER3, and so these are coming down the line.

 

Then for HPV-positive cancers, there is been interest in specifically targeting the viral immune response. There are a number of vaccine compounds that are moving forward in clinical trials. Then this very, very interesting molecule CUE-101, which we have worked with at Yale. This is a fusion protein where the HPV protein is settled in an HLA molecule with some IL-2 backpacks. This demonstrated very nice activity in the mid 40% range together with pembrolizumab.

 

[00:44:12]

 

Let us Return to an Earlier Case and Question

 

We have some time to return to our earlier case and question.

 

[00:44:16]

 

Case 1 Revisited: Patient With Metastatic HNSCC

 

This is my patient, 65-year-old man. Three years ago we treated him with chemoradiation for T4N2, p16+ oropharynx cancer. He did very well, but he achieved a complete response, but now he presents with bilateral lung nodules. We have proven that these are related to his p16+ oropharynx cancer.

 

[00:44:40]

 

Posttest 2

 

Would you treat him with:

 

• Cetuximab plus platinum plus 5-FU; I hope you guys are voting too up here;
• Pembrolizumab;
• Pembrolizumab plus platinum plus 5-FU;
• Nivolumab plus ipilimumab; or
• Dostarlimab.

 

The majority of you would go with pembrolizumab and chemotherapy. I think that makes good sense. It is also an option to give pembrolizumab. I think that that is also correct. I think some of the alternate chemo backbones would also be acceptable. I am curious to know how the 2 of you voted.

 

Dr Robert Haddad (Harvard Medical School): I put in pembro.

 

Dr Cristina Rodriguez (Fred Hutchinson Cancer Center): I did too.

 

Dr Burtness: Yeah. Well, in fact, that is what he got, and he did quite well. I will save some of his toxicities and on and off for Tina’s talk.

 

[00:45:59]

 

Posttest 2: Rationale

 

As we said, 2 correct answers here in the KEYNOTE-048 trial compared with chemotherapy plus cetuximab, pembrolizumab plus chemotherapy or pembrolizumab alone were superior in PD-L1 expressing patients. I did highlight that his PD-L1 expression was 9, which is a little bit on the lower side, and we potentially see a little bit more benefit with the addition of chemo in those patients.

 

With that, I am going to hand the remote over to Dr Haddad.

 

[00:46:38]

 

The Evolving Role of Neoadjuvant/Adjuvant Immunotherapy for Head and Neck Cancer

 

Dr Haddad: Thank you, Dr Burtness. Great pleasure to be here this morning. I will focus my talk on the curative setting, moving from the recurrent metastatic.

 

[00:46:56]

 

Locally Advanced HNSCC: Current Landscape in 2025 

 

A quick introductory slide about where we are today in 2025 in head and neck cancer locally advanced, previously untreated. We know about HPV as being a major cause of oropharyngeal cancer. We obviously continue to treat patients with chemotherapy and radiation. Cisplatin is the main therapeutic option we have for chemotherapy.

 

We have a new staging system as a few years back for HPV positive that is now different than HPV negative. As I mentioned, surgery, radiation and chemo are the mainstay of treatment for patients with head and neck cancer.

 

This is about to change we think this year with the first 2 positive phase III trials in the curative setting in head and neck cancer. One that we reported 2 weeks ago at AACR, the KEYNOTE-689. The second one is to be reported at the ASCO meeting, the NIVOPOSTOP.

 

[00:47:53]

 

Summary: Head and Neck Cancer in 2025

 

This is a summary slide of how we manage patients. Chemo RT is the standard. Some patients still benefit from induction chemotherapy or what we call it sequential chemo radiotherapy either for organ preservation or for very advanced disease in the adjuvant setting. Based on risk factors, we use radiation therapy or concurrent chemoradiation.

 

There is a move toward using weekly cisplatin now away from bolus cisplatin. In my opinion, both of these regimens are acceptable and are not getting into the debate anymore about bolus vs weekly. They are both totally reasonable. Patients who are not suitable for cisplatin, we have data to support docetaxel as an option with radiation or radiation therapy alone.

 

[00:48:37]

 

Concurrent/Adjuvant Immunotherapy 

 

Let us talk about the adjuvant, the concurrent immunotherapy with chemoradiation.

 

[00:48:43]

 

JAVELIN Head and Neck 100: Phase III Trial of Avelumab + CRT vs Placebo + CRT in LA HNSCC

 

This has been a very frustrating field with a number of negative trials. This is the first study we were involved with, and we essentially took the approach, if we are going to take “the bad head and neck cancer patients”, the HPV-negative larynx, hypopharynx, oropharynx, p16-. We are going to ask the question because those patients do not do as well as the HPV-positive, can we intensify their treatment by adding a checkpoint inhibitor, obviously based on the data from the recurrent metastatic disease that immunotherapy improved outcomes in those patients.

 

This is the first attempt we were involved in using a PD-L1 inhibitor, avelumab, with the standard of care cisplatin radiation. So randomized phase III study with the primary endpoint of progression-free survival.

 

[00:49:36]

 

JAVELIN Head and Neck 100: PFS

 

This unfortunately was the first negative trial in this space showing that the addition of avelumab to chemo and radiation did not improve progression-free survival. This did not proceed any further.

 

[00:49:49]

 

KEYNOTE-412: Phase III Study of Pembrolizumab + CRT vs Placebo + CRT in Locally Advanced HNSCC

 

The second big attempt to look at this is KEYNOTE-412 asking the question with a similar patient population. Think of your larynx, hypopharynx, HPV-negative patients who are looking for better options than chemoradiation. In this study, adding pembrolizumab to cisplatin radiation, again using the cisplatin 100 every 3 weeks with maintenance pembrolizumab. Primary endpoint of event-free survival.

 

[00:50:16]

 

KEYNOTE-412: EFS and OS

 

When you look at the results here for EFS and OS, this did not reach the statistical significance and was read as a negative trial, meaning that adding pembrolizumab to chemotherapy and radiation did not improve those 2 outcomes.

 

[00:50:31]

 

KEYNOTE-412: EFS and OS in Patients With PD-L1 CPS ≥20

 

When we looked further at the high CPS expressing groups, you actually see a trends of improvement with the addition of pembrolizumab to chemoradiation, but that was not the primary endpoint of this trial is to focus on those CPS 20 or more. As such, this was considered unfortunately to be a negative trial.

 

These 2 trials, close to 2,000 patients enrolled, were very disappointing for us in the field. We all started to feel that using immunotherapy with radiation does not make sense anymore. There is a lot of reasons to think why actually this would not work. People were starting to move away from the concept of using immunotherapy with radiation.

 

[00:51:14]

 

IMvoke010: Atezolizumab After Multimodal Definitive Therapy for High-Risk Locally Advanced HNSCC

 

The other study I was involved in very early in the process and based on the lung cancer data, PACIFIC study, we designed this trial with another Genentech PD-L1 inhibitor, atezolizumab.

 

Here we took different approach. We said we are going to take patients who have received curative therapy no matter what that curative therapy is. It could be surgery followed by radiation, it could be concurrent chemo radiation, it could be induction chemotherapy followed by radiation. As long as the patient gets into a complete remission, we are going to randomize those patients to the PD-L1 inhibitor or placebo.

 

[00:51:54]

 

IMvoke010: Outcomes

 

Unfortunately, this was a smaller study. It was less than 500 patients. You see here from the event-free survival, overall survival giving that PD-L1 inhibitor in an adjuvant setting did not improve outcome. A third negative trial in the curative setting for locally advanced head and neck cancer patients.

 

[00:52:17]

 

Summary: Immunotherapy in Locally Advanced HNSCC

 

This is a summary. There was another study called REACH from Europe using avelumab or cetuximab also was a negative trial. In the past, I would say 7 to 10 years and large number of negative trials adding chemotherapy radiation to a checkpoint inhibitor or given the checkpoint inhibitor as an adjuvant drug.

 

[00:52:38]

 

What About Neoadjuvant Immunotherapy?

 

This leads us to really what we saw 2 weeks ago, the neoadjuvant immunotherapy is giving obviously the checkpoint inhibitor before you touch the patient, before you give radiation, before you do surgery.

 

[00:52:52]

 

Rationale for Neoadjuvant Immunotherapy

 

Giving neoadjuvant immunotherapy as your first intervention. That is a rationale to using neoadjuvant immunotherapy. Obviously, there is a need to improve outcome in patients with high-risk head and neck cancer, those HPV-negative larynx, hypopharynx, oral cavity patients. We know this works. As we heard from Dr Burtness, this works in recurrent metastatic disease.

 

This might help induce an immune response early on before you have removed the lymph nodes with surgery or before you gave radiation, which essentially kills all the lymph nodes in the neck. Again, the neoadjuvant setting is ideal if you think about it. These patients have good performance status, they are untreated, lower burden of disease. Intact tumor allows you to induce a robust immune response.

 

A big question whether also down the road this could cause an immune reduction that might allow you to do, for example, a smaller operation, which would be what I think we will be talking about the next 10 years. Also very important point is if you give a neoadjuvant intervention, are you less likely to have high risk features after surgery? Thus, there will be a less need, for example, for chemoradiation and you will be more using radiation alone, thus decreasing the overall toxicity of the treatment.

 

[00:54:04]

 

Mechanisms of Action: Neoadjuvant Immunotherapy

 

Two mechanistic aspect we think are at play. The preexisting antitumor immunity is unleashed when you go with the neoadjuvant approach, and this induce a pathological response, which in most solid tumor, we know when you get this type of response, you are likely to improve outcomes with patients.

 

In this cartoon here, you see that you have these anti-tumor immune response that stick with the patients throughout the treatment and for a long time after that. That potentially is what is impacting the rate of distant metastatic disease and potentially local control.

 

[00:54:39]

 

Is Neoadjuvant Immunotherapy Safe Prior to HNSCC Surgery?

 

The first question we and others had to answer with this new advent approach, is it safe? Because obviously when you are doing something before surgery, the first question you get from our surgical colleagues, is this going to impact my operation? Is the patient going to have higher risk of post-op complication, intraoperative complications, flap failures, bleeding, infection.

 

The short answer is we and others did not see an increased surgical morbidity with during immunotherapy. All of these really endpoints that I mentioned in many of the trials that we and others have ran giving immunotherapy before surgery was safe.

 

[00:55:19]

 

Neoadjuvant Therapy for HNSCC

 

This is a summary of the trials that as of now have been published about neoadjuvant immunotherapy, the pembrolizumab 2 cycles. The first 1 really is the effort that was led by Dr Uppaluri, who works with me at Dana-Farber, start that effort at WashU, where he started with the concept of giving preoperative pembrolizumab and then we did the nivolumab-ipilimumab neoadjuvant trial. Then you have a number of trials there with atezolizumab and durvalumab.

 

All of these trials essentially gave us this notion that it is safe to do, it does induce response rates. It is not a massive response from a resist criteria. We know that when you just use immunotherapy. That really what led to the subsequent phase III trial that I am going to discuss.

 

[00:56:08]

 

Neoadjuvant Pembrolizumab for Locally Advanced HNSCC

 

The first effort that this is really what led to 689, these are the 2 studies that Ravi and Doug Adkins led from WashU and Dana-Farber starting with the first study where they gave only 1 dose of pembrolizumab and then did the surgery and then treat the patient with the standard of care. Then because this was safe, they saw responses. They decided let us give 2 cycles of pembrolizumab and see what happens. That is what is called cohort 2 here, was 2 cycles of pembrolizumab, 2 to 3 weeks before surgery, 28 patients. You do the surgery and after surgery treatment is the standard of care.

 

[00:56:42]

 

Neoadjuvant Pembrolizumab for Locally Advanced HNSCC: Dose/Timing

 

What they observed with these 2 studies, and that was really important. This is what led to 689 study. This was the basis for 689. But you gave 2 doses of pembrolizumab, that rate of pathological response was doubled. It went from 22% to 45%. It did make a difference to give 2 doses of pembrolizumab instead of 1 dose of pembrolizumab.

 

[00:57:06]

 

Neoadjuvant Pembrolizumab for Locally Advanced HNSCC: Outcomes

 

These are the outcomes shown here. But there was also noted here some down staging. What they observed is the patients who got neoadjuvant pembrolizumab were less likely to require chemoradiation after the surgery and more likely to require radiation therapy alone. There was a lower number of patients who had this extra nodal extension positive margin when they were treated with neoadjuvant intervention.

 

[00:57:36]

 

KEYNOTE-689: Perioperative Pembrolizumab + SoC for Locally Advanced HNSCC

 

As to 689, which again was presented 2 weeks ago at AACR in the plenary session. This is a global randomized phase III trial taking patients again with high-risk head and neck cancer, so your HPV-negative, larynx, hypopharynx, oral cavity patients. For HPV-positive is the bad HPV-positive, your T4 patients. Essentially, there was a very small number of HPV-positive patients on this trial.

 

The patients were randomized to either pembrolizumab 2 doses based on the data that I showed you before, followed by pembrolizumab with chemoradiation or pembrolizumab radiation based on the risk of the patient, whether they had extra capsular extension and positive margin, the 2 standard features we use currently to determine high risk vs low risk, and then adjuvant pembrolizumab.

 

The control arm is the standard of care. You go to surgery immediately and then you proceed with either radiation or concurrent chemo radiation.

 

[00:58:30]

 

KEYNOTE-689: Baseline Characteristics

 

These are the patient's characteristics. Close to 700 patients on this trial. Typical patient population. But I will point out that the oral pharynx group is small as you would expect. The biggest group is the oral cavity because these are the patients we operate on, right?

 

Larynx, hypopharynx were smaller groups and the majority of patients on this trial are the oral cavity patients. You see here that the vast majority of patients, 95% had a CPS of 1 or more. This is higher than when you saw with KEYNOTE-048 for example. But remember this is a patient population that is different than KEYNOTE-048. It is a previously untreated patient population. The vast majority of the patients on this trial had a CPS of 1 or more.

 

[00:59:12]

 

KEYNOTE-689: Treatment Disposition (ITT)

 

Here it gives you the treatment disposition. Two points I want to make. Patients were able to complete surgery in a similar rate between the 2 groups, so 88% and 87% were able to get their surgery. We can talk a lot about why some patients did not get surgery on the standard of care arm, for example. They changed their mind, they were start with unresectable, they got sick, a number of reasons. But the numbers were similar between the 2 groups.

 

The other point I want to point out here is the high risk features after surgery were 32% in the pembrolizumab arm and 44% in the standard of care arm. Giving neoadjuvant checkpoint inhibitor resulted in a lower number of patients needing cisplatin radiation. Again, raising that question of is that a down staging that is happening here that is allowing the patient to require radiation and not chemoradiation after their surgery?

 

[01:00:12]

 

KEYNOTE-689: Treatment Exposure (As-Treated Population)

 

This is the treatment exposure by cisplatin, by radiation. You see there was no differences between the standard of care arm and the pembrolizumab arm.

 

[01:00:23]

 

KEYNOTE-689: EFS in CPS ≥10 Population (Primary Endpoint)

 

Now the primary endpoint of this trial is event-free survival. there is a lot of data now to actually support that EFS is a marker of OS in head and neck cancer. That is the primary endpoint that was agreed upon with the FDA when this trial was designed.

 

The EFS in patients of CPS 10 or more favored the pembrolizumab arm. As you can see here, the median is 59 months in the pembrolizumab arm vs 26 months in the standard of care arm.

 

When you look at the distant control between the 2 groups, you see really a big difference of 10%. 6% in the pembrolizumab arm and almost 17% in the standard of care arm. It seems like giving neoadjuvant immunotherapy is impacting the rate of distant metastatic disease for these patients.

 

[01:01:12]

 

KEYNOTE-689: EFS in CPS ≥1 Population (Primary Endpoint)

 

When you look at the CPS 1 or more, again, we saw similar trends here. Remember that was 96% of the patients on the trial had a CPS of 1 or more. You see a significant difference between an EFS between pembrolizumab and between the standard of care.

 

[01:01:29]

 

KEYNOTE-689: EFS in All Participants (Primary Endpoint)

 

When we look at the all participants, again, similar trends here favoring pembrolizumab plus standard of care compared to standard of care. Across all the groups that were analyzed, and this was the statistical design of this study starting with 10, then 1, then all patients, this was a positive primary endpoint event-free survival.

 

The overall survival curves are favoring pembrolizumab but have not reached statistical significance yet. Those results will be presented later.

 

[01:02:01]

 

KEYNOTE-689: EFS by Subgroups in All Participants

 

Now when we look at the subgroups here, we see some trends. Oral cavity, for example, there is a benefit of pembrolizumab. Remember that was the majority of the patients on the trial. We do not see that trend in hypopharynx, which is a bad disease, very aggressive disease, but those numbers are very low.

 

Also, when you look at the stage III and stage IV, you see that stage IV, the hazard ratio go down to 0.8 compared to a much better hazard ratio of 0.4 in the stage III. A lot of these subgroups are going to be very informative for us going forward on how we apply this modality for our patients.

 

[01:02:36]

 

KEYNOTE-689: Investigator Conclusions

 

The conclusion from KEYNOTE-689, which we believe represent a new standard of care for patients with locally advanced head and neck cancer being treated with a surgical approach. There is an improvement in event-free survival for CPS10, CPS one and all comers. This is what I showed you. The overall survival is not reached yet and that continues to be analyzed. All of this is by how the study was designed.

 

This is now being reviewed by the FDA, and we are expected to hear from the FDA in the next 4 weeks about whether this would be a new approved indication for pembrolizumab in head and neck cancer patients.

 

[01:03:23]

 

NIVOPOSTOP: Adjuvant Nivolumab + SoC for Locally Advanced HNSCC

 

As I mentioned, there was another study that was reported by press release to be positive and that is the NIVOPOSTOP. This is a different study than KEYNOTE-689. In this phase III study, patients were treated with surgery first. A similar patient population of surgically treated head and neck cancer patients. After surgery, they were randomized based on their risk features to either nivolumab/chemo RT or chemo RT.

 

The high risk features using this trial are different than what we would use in the US because also there was inclusion of patients with multiple nodes and also perineural invasion, which are high risk features that in Europe are considered an indication to give chemotherapy with radiation.

 

This is a larger inclusion criteria for high risk than what we use in the US. Obviously, this trial does not have the intermediate or low risk patients that get radiation alone after surgery. This here is again be a plenary at ASCO in 3, 4 weeks. We are looking forward to see the data from this trial so that we have a better assessment of how this treatment is performing compared to what I showed you with KEYNOTE-689.

 

[01:04:38]

 

NIVOPOSTOP: Adjuvant Nivolumab + SoC for Locally Advanced HNSCC

 

There are 2 ongoing phase III trials currently asking a question of adjuvant immunotherapy. Remember in the first slide I showed you the IMvoke trial, which was a PD-L1 inhibitor. But again, as I mentioned, IMvoke included all patients, surgical, non-surgical. These 2 trials are including patients who are treated with chemoradiation, high-risk patients. In the first one, which is the JADE trial, which Dr Burtness is leading, is essentially comparing PD-L1 from GSK vs placebo with an event-free survival as primary endpoint.

 

The second trial, eVOLVE trial from AstraZeneca is essentially taking a similar patient population but testing a bispecific CTLA-4 PD-1, again after chemoradiation volrustomig vs placebo vs observation in this trial after chemoradiation.

 

These 2 trials are ongoing and again, asking a question. It is a different patient population than KEYNOTE-689, which is a surgical population. This is a nonsurgical population. Think of all your patients who are getting chemo RT currently for their treatment, those are the patients who would enroll on 1 of these 2 trials.

 

[01:05:51]

 

Conclusions

 

My final slides. We know that the standard of care for head and neck cancer patients, the curative setting unfortunately has not changed in the past many decades. It has remained surgery, chemotherapy and radiation. What I showed you, I hope convinced you that we are about to see a massive change in how we treat these patients based on KEYNOTE-689 and the role of neoadjuvant immunotherapy. This is really going to bring a number of issues to the front, and most important one is the collaboration in the MultiD clinic and bringing the medical oncologist early on to the conversation for those surgically treated patients so that the appropriate patients can be treated with a neoadjuvant intervention before their operation.

 

We await the results of NIVOPOSTOP, but we have the results of 689 already. No matter what the outcome would be, I would expect that in the next 6 months you will see a massive change of how we treat head and neck cancer patients in the US and worldwide.

 

Thank you very much.

 

[01:06:55]

 

Let us Return to an Earlier Question

 

Posttest 1

 

The question that Dr Burtness has asked in the beginning and based on the phase III KEYNOTE-689 in patients with locally advanced head and neck cancer, for which population of patients there was an EFS benefit with the addition of perioperative pembrolizumab to standard of care surgery and radiation with cisplatin? Is it:

 

1. CPS more than 20;
2. CPS more than 10;
3. CPS more than 1; or
4. All patient regardless of their CPS level.

 

[01:07:48]

 

Posttest 1: Rationale

 

Okay. Most of you got this correctly. Based on the data that I showed you, the outcome was seen in all patients regardless of the CPS score. This is the explanation.

 

There was 1 question from the audience. Do you want to read it, Dr Burtness?

 

Dr Burtness: Without mature overall survival data, would you consider a KEYNOTE-689 approach for locally advanced disease now? Maybe kick it to Tina first and then you, and then all 3 of us can opine.

 

Dr Rodriguez: Well, I mean I always think back to why we do neoadjuvant therapy, and I think that the data, at least from the Mackenzie[?], suggests that it really impacts distant disease-free survival.

 

That was what was seen in KEYNOTE-689. All of us who treat these patients know that that is really an outcome that concurrent administration of systemic therapy has not really had a major impact on. I do think it is a clinically relevant endpoint, and I do think that it makes a difference for our patients.

 

Once it is approved, I think our practice will change to incorporate a neoadjuvant pembrolizumab.

 

Dr Haddad: I feel the same. I think an event in head and neck cancer patients is a terrible outcome. Event-free survival is an important outcome in head and neck cancer based on a number of studies. When I look at the curves that I showed you, if I am the patient, I want to be on the top curve, not on the lower curve.

 

For me, this is a practice changing study. Obviously, we continue to analyze. There would be a lot of analysis coming from this trial including OS, which I did not show the slide, but the trends are favoring pembrolizumab as of now. It has not reached statistical significance yet. Those numbers will be available, would be reported. But if you ask me today what I would do pending FDA approval, this would be what I would offer my patients.

 

Dr Burtness: I am the same. I think it bears thinking about why the FDA, which tends to be conservative in the curative setting, has agreed to accept event-free survival in head and neck cancer. As Robert said, an event is often life shattering, right? You lose your tongue, you lose your larynx, you are back on chemo, you are out of work for a long time.

 

The head and neck international group has done a big survey of use of event-free survival and finds a very large number of regulatory bodies internationally also accept this. This wasn't a foregoing conclusion that this study would read out even for event-free survival at the first interim analysis. We are hearing about it now because there is such a dramatic difference at the first interim analysis. You would not really expect with the majority of the patients, even on the control arm still being alive at this interim analysis, you would not expect to see the OS signal yet.

 

In my mind, as long as we are not seeing a deleterious OS signal at this point, the small number of patients who do not make it to surgery or whatever, but the OS trend is strongly favoring the neoadjuvant arm. But at a time when more than 50% of the control arm patients are still alive because it is a first interim analysis, I think that it is reasonable to know that we have to wait for that.

 

I think we maybe have a second for 1 more question. Lovely results, a clear impact of neoadjuvant therapy based, as you say, on the expectation of hitting microscopic distant disease, exploiting the fact that the tumor draining lymph nodes are still present. I think it must be said, a largely biomarker selected patient population, right? 95% were PD-L1 expressing, and over 95% were HPV negative. But we still are seeing 3-year results under 70%.

 

I guess although it looks lovely to say, oh, now we can cut back on the adjuvant therapy. Given that we aren't seeing the big difference in local regional control, do you think that we still need to ask the question of should we be continuing the radiosensitizer for the responders? You see what I am saying?

 

Dr Haddad: Yeah. The question is whether keeping that platinum in those patients would give us even a better? It is a tough question to answer obviously, because we do not have the data.

 

Dr Burtness: No, I think it would have to be a study.

 

Dr Haddad: Right. There would be a study. To me, I think the study that I think all of us are itching to do is to add something to pembrolizumab in the neoadjuvant setting to improve those NPRs, to improve those responses, getting to do maybe a lesser operation for patients that will have less impact on quality of life. That is 1 question that Dr Burtness is raising, that would be one of another study to ask to see if we can bring even those numbers higher.

 

But again, this really brings up a number of other questions like how can we improve those data by adding a drug or multiple drugs to the neoadjuvant? But also we have then to contend with the notion, is this going to impact surgery? Because everything you do in neoadjuvant, the first question you ask is going to impact your intervention, which is the curative intervention here, surgery.

 

If you think of head and neck, many of us before this study read, we are giving patient pembrolizumab, carbotaxol before their surgery for the bad head and neck cancer. If you talk to a lot of people here in this meeting about what do they do in the neoadjuvant setting? It is amazing to me how many people were giving chemo IO in the curative setting before surgery and the bad had a neck cancer. This is not your T1, T2. This is your T4 that needs a total glossectomy or a near total glossectomy.

 

The patient does not want the operation. They are technically resectable but not really. People do not give pembrolizumab alone. They give chemo pembrolizumab for that patient. I have done it many times. Barbara, I know did that many times. This is how we manage those patients with the bad head and neck cancer because you will get significant debulking when you add chemo to the immunotherapy.

 

A lot of what you are going to hear the next 10 years is how can we improve on the data that I showed you? How can we make this better than what I showed you? It is going to involve a number of interventions. It could be different adjuvant intervention, different neoadjuvant intervention. The piece I am the least excited about in 689 is the chemo RT portion and pembrolizumab because I do not know how much that contributing.

 

I am excited about neoadjuvant piece and the adjuvant piece, but chemo RT, I am lukewarm to it. Because we know from prior studies adding pembrolizumab to chemo RT did not make a difference. But the intervention here started early and maybe that is really where you are impacting, where you are getting that EFS distant metastasis control.

 

The outcome really is related. If you look at all what I showed you, 412, IMvoke, JAVELIN, the difference here is that here we started the intervention before we touched the patient, before we did anything. No radiation, no surgery, just neoadjuvant first thing you do to the patient. That is what is really I think driving all this benefit.

 

Dr Burtness: But where we are still lagging is local regional control.

 

Dr Haddad: Yes.

 

Dr Burtness: I am not saying that we know the answer. But it is worth asking.

 

Dr Haddad: More work.

 

Dr Burtness: Now, Dr Rodriguez will talk to us about recurrent/metastatic nasal pharyngeal cancer and then as well, we do need to address management of immune-related toxicity.

 

[01:16:21]

 

Leveraging Immunotherapy to Optimize Treatment of Recurrent/Metastatic Nasopharyngeal Cancers

 

Dr Rodriguez: Thank you so much for having me. What a pleasure to be here. I was reflecting while sitting in the audience about how I was here 5 years ago doing this talk with Dr Burtness and Dr Ferris[?]. I was just amazed at how much progress and how much work has been done in going from the initial results of KEYNOTE-048 and now bringing immunotherapy, the curative intent setting.

 

With that, I am here to talk about the role of immunotherapy in a less common subset of head and neck cancer that we see here in North America, nasopharyngeal carcinoma.

 

[01:16:57]

 

Case 2: Patient With Recurrent Middle Ear Effusions

 

I will start with a case. This is an actual patient of mine, who was a 49-year-old gentleman who is Cantonese speaking. He comes from the Guangdong region, which has the highest incidences of nasopharyngeal cancer in the world. He had recurrent left middle ear effusions and on physical exam he had a tympanostomy tube in the left eardrum and palpable left-sided level 5 adenopathy, which is a unique first echelon set of lymph nodes for nasopharyngeal cancer.

 

I obtained imaging and there is a mass in the left fossa of Rosenmüller explaining his middle ear effusion.

 

[01:17:36]

 

Case 2 Continued: Patient With Recurrent Middle Ear Effusions

 

A biopsy of the left level 5 lymph node was obtained. On his the pathologic findings, there is an undifferentiated non-keratinizing carcinoma noted. There is a brisk lymphoid infiltrate around it. When EBER in situ hybridization was performed, it was very positive. We staged him with a PET CT. Unfortunately, he has extensive bone and liver metastases. However he's healthy. His labs are normal. His ECOG is zero.

 

[01:18:07]

 

Pretest 3

 

The question for the audience, for this patient, what would you recommend:

 

1. Chemoradiation with cisplatin;
2. Gemcitabine plus cisplatin times 3, followed by chemoradiation; and
3. There is a few chemo options here, including gemcitabine and a PD-1 inhibitor.

 

Okay, so it looks like most people recommend platinum gemcitabine and a PD-1 inhibitor. We are going to talk about the data that supports this in the next few slides.

 

[01:19:01]

 

Recurrent/Metastatic NPC

 

Recurrent metastatic nasopharyngeal cancer is a very unique disease. Epidemiologically, it has got a peri equatorial distribution.

 

[01:19:08]

 

Recurrent/Metastatic NPC and PD-1 inhibition

 

Most of these patients are treated with curative intent, but a significant proportion of them will recur. It is a chemo-sensitive disease. There is multiple active systemic agents. A lot of the data generated in this disease is really through the efforts of academic centers in endemic regions, particularly of Sun Yat-sen University.

 

This is true of this study, which established gemcitabine and cisplatin as the favored chemotherapy combination in the first-line recurrent metastatic setting. When it was compared to 5-fluorouracil and cisplatin, there was a statistically significant PFS and OS benefit to the combination and a fairly impressive overall response rate of 64%.

 

Now, PD-1 inhibition is an approach that is particularly relevant in the nasopharyngeal cancer population. Why is that? Because there is near universal PD-1 expression, especially in endemic disease. This disease is also called the lymphoepithelioma and it is characterized by a robust tumor immune cell infiltrate. This is why multiple anti-PD-1 monotherapy studies that were explored in this disease actually showed activity somewhere between 20% or 30% of objective response rates.

 

You are probably familiar with nivolumab/pembrolizumab, but the 3 PD-1 inhibitors in this chart are all China developed PD-1 inhibitors. The camrelizumab experience was in a basket study, where an arm in that study combined camrelizumab with the gemcitabine-cisplatin combination that we talked about.

 

In a small population of patients, there was a very impressive overall response rate of 91%, which led to evaluation of this triplet combination in phase III setting.

 

[01:21:11]

 

Systemic Therapy Recommendations for Nasopharyngeal HNSCC

 

This is the current, would likely to change in the very near future, NCCN recommendation for nasopharyngeal cancer. Particularly in the first-line, the regimens recommended are cisplatin/gemcitabine based, and the combination of PD-1 inhibitors are recommended.

 

[01:21:34]

 

JUPITER 02: First-line Toripalimab + Gemcitabine/Cisplatin for R/M NPC

 

The first landmark study that changed the standard of care for nasopharyngeal cancer is the JUPITER-02 trial, again run by investigators from Sun Yat-sen University. This was a randomized, double-bind, phase III trial in first-line recurrent metastatic nasopharyngeal cancer.

 

The design was one with PD-1 inhibitor. Toripalimab was added to gemcitabine and cisplatin and then used as maintenance after 6 cycles. The control arm received placebo. The primary endpoint for this study was progression-free survival by blinded independent radiologic review. Secondary endpoints are listed on this slide.

 

[01:22:18]

 

JUPITER-02: PFS

 

This was a positive study. It met its primary endpoint. There was, as you can see, a significant improvement in progression-free survival for patients who received the PD-1 and chemo combination. Not shown in this slide was this study being 1 of the earliest phase III trials actually showed a statistically significant overall survival benefit, which was a secondary endpoint. That is a pretty impressive hazard ratio for PFS of 0.52.

 

[01:22:51]

 

JUPITER-02: PFS Subgroup Analysis

 

Subgroup analysis of this data shows that the majority of the different patient subsets all seem to benefit from the addition of toripalimab to chemotherapy.

 

[01:23:07]

 

Recent FDA Approval of Toripalimab for NPC

 

In October of 2023, the FDA granted approval for toripalimab in combination with chemotherapy as first-line treatment for patients with metastatic nasopharyngeal cancer. Based on a phase II trial in previously platinum chemotherapy treated patients that was also approved as a single agent following platinum chemotherapy.

 

[01:23:35]

 

Additional Phase III Clinical Trials in First-line R/M NPC

 

There were 2 other trials that followed the JUPITER study, and these 2 again were quite similar in design. One was CAPTAIN-1ST and the other was RATIONALE-309. As you can see, both of these studies accrued the same population, first-line recurrent metastatic. Both of these studies were run primarily in endemic regions. Both of them randomized patients between a chemoimmunotherapy regimen vs placebo plus chemotherapy.

 

RATIONALE-309 was different in that it allowed for crossover in patients who have progressed. These studies similarly had a primary endpoint of PFS.

 

[01:24:20]

 

Key Phase III Trials: Patient Characteristics

 

Now, this table compares these 3 studies that were reported in very short timeframe one after the other. But as you can see, there is a fairly equal distribution of ECOG Performance Status 1, recurrent disease, and PD-L1 positive patients. The CAPTAIN-1ST did not test for PD-L1 and that was not reported.

 

When they looked at the results of particularly RECIST 1.1 response rates among these agents, there were fairly similar observations. The 1 striking thing about this table is if you look at the JUPITER-02 trial. The median duration of response is 18 months in the paper compared to 8.5 in both the CAPTAIN-1ST and RATIONALE.

 

I think this is an artifact of the fact that the data is more mature in JUPITER-02. The median follow up there was 3 to 4 times longer than what was in CAPTAIN-1ST or RATIONALE.

 

Regardless, because of the maturity of the JUPITER-02 trial, there is a statistically significant benefit from an overall survival perspective in that study, which we anticipate will probably observe in the subsequent studies as well.

 

[01:25:44]

 

Key Phase III Trials: Toxicity

 

From a toxicity perspective, I think when we combine chemotherapy with immunotherapy, we tend to see a much higher grade 3 or higher adverse event profile. This was something that was similarly observed in these 3 trials where we saw high grades of grade 3 toxicities attributable more to the chemotherapy than immunotherapy. Because if you look at the immune-related treatment adverse events that were high grade, they were similar to what we observed with pembrolizumab alone.

 

[01:26:18]

 

AK105-304: First-line Penpulimab + Chemotherapy for R/M Nasopharyngeal Carcinoma

 

Late to the party, but still was successful was this fourth, very similar in design, randomized phase III trial in nasopharyngeal carcinoma that was presented 2 weeks ago at AACR. This study, once again, once a chemo immunotherapy vs placebo and chemotherapy in the first-line setting with very similar primary endpoints but using a PD-1 monoclonal antibody called penpulimab.

 

[01:26:52]

 

AK105-304: Outcomes

 

This was also a positive study with an impressive hazard ratio favoring the chemoimmunotherapy arm in terms of PFS, with a median duration of response that was very similar to CAPTAIN-1ST and RATIONALE-309, and very similar AEs reported in the prior 3 randomized studies.

 

[01:27:16]

 

Recent FDA Approval of Penpulimab

 

As of last month, this has also joined toripalimab as a first-line FDA approved option, chemoimmunotherapy in first-line setting, and also based on earlier phase II trial showing activity in previously chemotherapy treated patients. It was also approved in patients previously treated with platinum.

 

[01:27:40]

 

Let us Return to an Earlier Case and Question

 

Case 2 Revisited: Patient With Recurrent Middle Ear Effusions

 

With that, let us go back to the patient with recurrent middle ear infusions who has an EBER positive non-keratinizing nasopharyngeal cancer.

 

[01:27:52]

 

Posttest 3

 

What would you use among these 6 options? 100%. All right.

 

[01:28:29]

 

Posttest 3: Rationale

 

Well, the answer is E, which everybody got, and again, there is multiple randomized phase III data suggesting that chemoimmunotherapy with a PD-1 inhibitor is a standard of care in the first-line setting supported by landmark studies.

 

[01:28:47]

 

Ongoing Investigations in R/M NPC

 

Okay, so where are we going from here? Well, there is a tremendous amount of interest in improving upon what we now know in the recurrent metastatic setting. Some of the themes of investigation in this disease include maintenance therapy combinations, chemo immunotherapy in the non-endemic nasopharyngeal populations and post-chemo immunotherapy regimens.

 

[01:29:12]

 

REMAIN: Maintenance Nivolumab + Relatlimab for R/M NPC

 

I am going to go through a few trials that are open in the United States that are currently accruing and trying to answer these questions. One of them is the study called REMAIN, run by NRG. This is a randomized open-label phase II trial looking at patients with EBV-associated recurrent metastatic nasopharyngeal cancer.

 

They will receive 6 cycles of chemoimmunotherapy and patients who have not progressed will be randomized to an intensified arm looking at nivolumab with relatlimab, a LAG-3 inhibitor vs nivolumab. The primary endpoint of this study is PFS, and it is currently accruing.

 

[01:29:48]

 

Nonendemic NPC

 

There is also increasing interest, especially here in North America where we do not see a lot of endemic nasopharyngeal cancer. There has been a few reports, especially in the Midwest where the immigrant population is not as large as on the coast, but these single institution studies suggest that in North America there is a significant proportion of patients who have either HPV-positive, EBV-negative disease, or EBV and HPV-negative disease. Their outcomes do not appear to be as good as the endemic populations, which begs the question, is our current standards set in the endemic setting applicable to this population?

 

[01:30:36]

 

TRANSPARENT: First-line Toripalimab + Platinum + Gemcitabine for R/M Nonendemic NPC

 

That is a subject of a currently enrolling trial called TRANSPARENT. This is also run by the NRG, but it is looking at the established combination of toripalimab-platinum and gemcitabine in patients who are either EBER and EBV-negative and HPV-positive or keratinizing nasopharyngeal cancer. Essentially the population that was not enrolled in these large phase III trials, with a primary endpoint of overall response rate.

 

[01:31:08]

 

A092105: Nivolumab + Ipilimumab ± Cabozantinib for Advanced NPC Progressing After Platinum + ICI

 

Now what to do if patients progress on chemoimmunotherapy? That is trying to be addressed by this trial run by the Alliance Group. This is a randomized phase II trial looking at patients who progress after chemoimmunotherapy and randomizing them to a triplet combination of nivolumab, ipilimumab or cabozantinib vs nivolumab or ipilimumab with the primary endpoint of PFS.

 

[01:31:34]

 

Encouraging Combinations in Second-line Setting

 

There are some very encouraging combinations in the post chemo IO setting, particularly those that involve combinations with VEGF inhibiting TKIs such as lapatinib and erlotinib. There is some data that HDaC inhibitors plus antivirals may be effective and the EGFR antibody drug conjugates also look to be quite encouraging in efficacy.

 

[01:32:05]

 

Moving ICI Therapy to Curative Intent Setting

 

Of course, similar to what happened in more common mucosal head and neck cancers, there is significant interest in moving the immune checkpoint inhibitors into curative intense setting, which currently for high-risk disease involves induction gemcitabine and cisplatin followed by chemoradiation.

 

There are now an accumulation of early phase clinical trials suggesting that this is a very encouraging approach in this population. So lots to be expected in the near future regarding nasopharyngeal cancer.

 

[01:32:40]

 

Safety Considerations With Immunotherapy

 

If there are no questions, I think I will move forward. Managing toxicities related to immunotherapy.

 

[01:32:49]

 

Case 3: Patient With Immune-Mediated Pneumonitis

 

We will start with the case again. Again, 1 of my patients. This is a gentleman with p16+, recurrent metastatic squamous cell carcinoma with a CPS of 50 that I started on single agent pembrolizumab. Right before cycle 4, which is 12 weeks in, he reports a mild cough. He hikes a lot, but all of a sudden now maybe he just gets a little short breath when hiking, but performance status is zero.

 

I repeated a CT chest, and his disease has had significant reduction in size including his neck nodes. However, I saw these new pulmonary infiltrates. We talked to radiology and pulmonary medicine, and they feel like this looks like pneumonitis.

 

[01:33:34]

 

Pretest 4

 

Currently what would you do for this patient? Would you:

 

1. Continue pembrolizumab and treat with steroids;
2. Hold pembrolizumab, treat with steroids, and then rechallenge;
3. Hold, treat with steroids and then reduce the dose; or
4. Permanently discontinue pembrolizumab.

 

Okay. Most people said they would hold, treat with steroids and rechallenge.

 

[01:34:13]

 

Toxicities With HNSCC Chemotherapy, Targeted Therapy, and Immunotherapy

 

When we talk about toxicities with systemic treatments, the introduction of immune checkpoint inhibitors really introduces a very different type of toxicity profile that we see among our patient population. The toxicity profile is one of autoimmunity or inflammation, and it can involve various organ systems. These manifestations can be quite wide ranging depending on what organ system is affected.

 

[01:34:50]

 

PD-1 Inhibitor Time to Onset and Time to ResolutionBased on Melanoma Data

 

This is a great paper that I like to quote. This is from an experience in melanoma patients receiving single-agent nivolumab. There is about 500 patients in this paper, but this paper really informs us when we tend to see these toxicities. The most common toxicities regardless of grade are the one in the chart on the left side of the screen, which is we tend to see a lot of skin toxicity and we tend to see it early on.

 

It is usually like pruritus or very mild rash. We also can see GI toxicities a little before 10 weeks into treatment. But the less common toxicities that can be severe and can be life-threatening are really the hepatic, the pulmonary side effects. We can see that in terms of timing, they are a little bit later than skin toxicity and usually occur around that 10 to 12-week period after starting treatment.

 

[01:35:59]

 

Immune Checkpoint Blockade: Patient Assessments

 

It is really important to have a high level of suspicion when seeing these patients while they are on treatment. It is important to ask them about potential for autoimmune AEs. We tend to include labs that encompass endocrine labs, thyroid, cortisol. We look at pancreatic enzymes. It is also important to think about when we usually get baseline imaging before treatment and about 3 months after treatment, but sometimes we need to do these scans more frequently, especially if patients are complaining about shortness of breath or cough or fever. That makes us worried about pneumonitis.

 

The other point to make with this slide is these can occur even after you stop immunotherapy. It is important to keep that in mind.

 

[01:36:49]

 

General Guidelines for Management of Immune-Related AEs

 

In general, the grading of immune-related AEs, most people are familiar with this. For grade 3 and grade 4, which are life-threatening, it is pretty obvious that you have to stop treatment or for very asymptomatic patients we know we can continue treatment. Where it gets dicey is really that grade 2 moderate symptoms. My rule of thumb is if I am concerned enough about a potentially life-threatening immune checkpoint inhibitor side effect, if I am even thinking about holding, I usually hold.

 

[01:37:24]

 

Immune-Mediated Endocrinopathies: Symptom Management

 

Endocrinopathies. These are pretty common, especially in the head and neck population who have either had radiation therapy to the thyroid or are already hypothyroid. It is very frequent for us to see a slight reduction of the TSH followed by hypothyroidism afterwards.

 

The NCCN guidelines recommend an endocrinology consult. Personally, we do not do this at our institution because of the number of patients who are hypothyroid, but it is fairly easy to manage with levothyroxine replacement. Hypophysitis is something that can result in a smaller population of patients. We usually involve endocrine for this. It is usually manifested by a headache. Usually, these patients require hormone replacement therapy.

 

[01:38:17]

 

Endocrine irAEs

 

Again, in endocrine AEs generally what is needed will be hormone replacement. In general, we tend to continue PD-1 blockade with appropriate replacement treatment.

 

[01:38:35]

 

Immune-Related Colitis

 

Immune-related colitis. That is something that it can be pretty dramatic in its presentation. Patients will complain of abdominal discomfort and pretty much unremitting diarrhea. Like many institutions, there is usually services within the institution such as gastroenterology that tends to respond to these immune checkpoint-related toxicities. We generally diagnose them through colonoscopy and biopsies. These patients do require steroid treatment.

 

[01:39:12]

 

Management of Grade 3/4 Immune-Related Colitis

 

In patients who respond nicely to steroid therapy, you can consider resuming anti-PD-1 therapy, but patients with life-threatening toxicity, you really have to stop immunotherapy and discontinue. There are patients that require inpatient care because of dehydration and the need for aggressive IV steroids.

 

There are patients who can be refractory to steroids and who require biologics such as infliximab and vedolizumab. Really working hand in hand with our gastroenterology colleagues is very important for these patients.

 

[01:39:50]

 

Biopsy-Proven Pneumonitis Is Highly Variable in Presentation

 

Like I said, pneumonitis can happen in a small proportion of patients. It can have various appearances on CT imaging. Again, we have pulmonary colleagues who are really very useful to work with in terms of diagnosing these and managing patients.

 

[01:40:08]

 

Management of Immune-Mediated Pneumonitis

 

Especially, if it is grade 2, I think it is important to recognize you have to hold treatment. You have to be pretty aggressive with treating these patients with steroids. This is a potentially life-threatening toxicity. The small amount of treatment-related deaths in the trials of immunotherapy are really in patients who develop pneumonitis.

 

Again, you can escalate treatment to IV steroids or with a more intense immunosuppression if they do not improve again with inpatient admission and working with pulmonology.

 

[01:40:49]

 

Let us Return to an Earlier Question

 

Let us return to the question.

 

[01:40:54]

 

Posttest 4

 

Pneumonitis. After checkpoint inhibitors, what should we do?

 

1. Continue, treat with steroids;
2. Hold, treat with steroids and rechallenge;
3. Hold, treat with steroids and rechallenge at a reduced dose; or
4. Permanently discontinue.

 

Okay. What actually happened to this patient was when all the infiltrates in his lungs resolved, I talked to him about rechallenge, he did not want to. I am proud to say though that he is 4 years out of the last dose of pembrolizumab, and he is hiking. He does not have any evidence of progressing disease, and he is happy as a clam.

 

[01:41:47]

 

Posttest 4: Rationale

 

Everybody got that question correct.

 

[01:41:56]

 

Poll 3

 

Before we go to questions, we wanted to get a poll from the audience about how effective this CME program was. Do you plan to make any changes in your clinical practice based on what you learned in today's program? We are not going to show the results of these polls.

 

[01:42:32]

 

Poll 4

 

Okay. Please enter what that change would be.

 

Okay. I am going to hand it back to Dr Burtness to end the event.

 

[01:42:55]

 

Q&A

 

Dr Burtness: Okay. We just have a few minutes left, but there have been some interesting questions that have come in. Some of them sort of tended in the same direction. I am going to conflate these 2 questions. It has to do with the fact, if the patient got immunotherapy before they went to surgery and then they are in that 35% that ultimately recurs anyhow, what is your first-line therapy going to be? Please let your answer be influenced both by what you expect the pattern of recurrence to be and when you expect those recurrences to happen.

 

Dr Haddad: No, I mean I think this is a great question because this is something that we expect that KEYNOTE-689 will profoundly impact the first-line recurrent metastatic setting, which is currently where we use pembrolizumab. To me, if the patient has gone through a 689 regimen and they have recurrent disease or metastatic disease immediately after stopping treatment, the question becomes do you rechallenge the patient with pembrolizumab at that time since they were responding to it before and they progressed only when you stopped it.

 

I have done this already on the KEYNOTE-048 patients, and there the experience has been not so great. Meaning when I stop the drug and the cancer comes back and I rechallenge, I do not see the same profound response that I initially seen.

 

I think when we think about platinum resistance and when do we challenge patients with platinum, we always arbitrarily use the 6 months timeframe to say someone can be challenged with cisplatin. Well, we have to look at this in more detail, but there are a number of trials right now adding a second drug to pembrolizumab. We heard about the EGFR inhibitors, the VEGF inhibitors, all of those.

 

I think we are reasonable to rechallenge the patients with a checkpoint inhibitor for those patients, but likely it will be in addition to another second drug based on the currently ongoing phase III trials. But this would be really something we will have to think about.

 

Dr Burtness: Do not you expect that many of them are going to be locally salvageable? I mean you look at what your rate of distant metastasis and distant plus local was, that is only what, about 3%?

 

Dr Haddad: No, I am assuming they do not have a curative option. That is important because if someone has a local recurrence that is resectable, obviously this is where we are going.

 

Dr Burtness: I mean my read of the data was that there was a pretty substantial subset, right? Because of local only recurrence.

 

Dr Haddad: Yeah. Remember all of these trials, the assumption when we do the recurring metastatic trial of patients, the assumption is there is not a curative option for those patients. If there is a local recurrence and that if it is a larynx patient, you can do a laryngectomy, and you are going to treat that patient even though those patients would have had surgery.

 

We look at reradiation as an option for some patients. We look at surgical resection. We have looked at things like radiofrequency ablation or cryotherapy. All of these would still be hold through. I was more focused on those patients who do not have those low.

 

Dr Burtness: I mean I think it will be interesting as we see more data, but it seems to me the distant recurrences was actually the smallest subset.

 

Dr Haddad: But I am curious to hear your opinion as the KEYNOTE-048. How do you approach now the first-line incurable with no local therapy options? What do your cutoff?

 

Dr Burtness: In my mind, there is a question about how long they were IO free. I think to Tina's point that sometimes the patient's ready to go back on immunotherapy after an immune-related toxicity, but their scan looks so good, and their memory of the high dose steroids is so fresh that it does not happen. Some of those patients, it is 2 years. I mean some of them are cured like yours. Some of them it may be 2 years before you have to rechallenge.

 

I have to say I do think that those patients who've been off for a bit, we see nice responses. It is also I would like to bring out though from KEYNOTE-048, Kevin Harrington had a paper in the JCO, where he looked at something called PFS2. The patients had progressed on pembrolizumab mono on KEYNOTE-048. Then what was their second line therapy that they went on and how did they do on that? Obviously, it was a lot of chemo and cetuximab.

 

But if you looked at the duration of control in that second period of progression, the fact that they had previously had IO made them more responsive to chemotherapy and the duration of that PFS2 was higher for the patients who had first had immunotherapy than for those who had not. I think that is very consonant with what Robert was talking about before. You are expanding the population of existing anti-tumor immune cells, and they remain in the patient over a very long period of time, which is why we see the long tail on the curve with immunotherapy trials.

 

I think that even if you have to give them chemo cetuximab because they just came off pembrolizumab, I would anticipate that the fact that they previously had the immunotherapy exposure would lead you to expect a higher response rate and duration of response to that therapy. One of the reasons Kevin did that analysis is we had seen from the Institut Curie in Paris a report that patients going back on chemo they had seen before, after they have had IO had better responses.

 

Then I know we are a little over time, but if people want to hear the answer to 1 more question, and I think this is something that we talked about a lot at ECOG this week. Do you think both the adjuvant and neoadjuvant pembrolizumab doses are important in KEYNOTE-689? Maybe I will turn to you, Tina.

 

Dr Rodriguez: Well, I think it is fair to say we do not know the answer to that question. To me at least my view of the data is that Bob mentioned, giving it before or after curative intent therapy seems to make most sense for reasons that we do not understand well. Giving it concurrent with chemoradiation does not appear to impact outcome, but there is something about giving it before or after that makes a difference for patients.

 

Dr Burtness: Yeah, I mean people may know this experiment that was done by Silvio Gutkind at UCSD where he had 2 sets of mice, 1 got a neck dissection, 1 got a sham neck dissection, and then they were exposed either to a PD-1 inhibitor, or CTLA-4 inhibitor, and there was no response to the immunotherapy in the neck dissected mice. That is 1 reason I think the neoadjuvant helps.

 

Another is the response rate that we see. The depth of the response can be quite profound.

 

I think maybe the person who asked this question was thinking ahead to NIVOPOSTOP. One of the things that I think I have been thinking about a lot is if we are equally convinced by both studies, which therapy will we lean towards? I think what concerns me about the NIVOPOSTOP is not only what we have been saying about what we think is happening biologically with the early exposure to the immune checkpoint inhibitor, but that was confined to the high risk patients.

 

But as Robert showed you in KEYNOTE-689, the stage III patients who I think probably generated more of the margin negative non-E&E patients were the group that derived the largest benefit, and they would not be offered an immune checkpoint inhibitor if you used only the high-risk post-op approach.

 

I do not know what the 2 of you think about that? Then we will wrap up.

 

Dr Haddad: I completely agree. I mean, I think what will happen now is the question about adjuvant because this is how the study was done. I mean, this is what I think should be used until a new study is done that is trying to tease out the benefit of adjuvant. In my practice, the patient I decide to give neoadjuvant treatment for, that patient will essentially follow a 689 regimen with, because that is how really the study was done. We do not have the data right now to say that, yes, those 2 cycles of pembrolizumab, albeit essential, are the only ones that needed to happen until we have more data.

 

I think the adjuvant phase should be part of the treatment from now and the NIVOPOSTOP, I agree with everything you have said. I mean by doing that intervention post-op, then you are not treating your intermediate low risk patients. Plus whether the down staging is real, which I think it is, you are giving less cisplatin for patients on the neoadjuvant.