Neoadjuvant IO before resection in HNSCC and implications for RT

[Palex80]

---

Members do not see this ad.

Hey!

Our med oncs are starting to push for perioperative Pembro in resectable HNSCC based on this trial.

https://www.nejm.org/doi/full/10.1056/NEJMoa2415434

What are the implications for adjuvant RT?

The trial treated pretty much everyone with at least 60 Gy. I am wondering what am I supposed to in patients achieving major pathological response. Do I still treat the same way?

Have you seen this happening at your institutions?

 

[drowsy12]

Treat per path indications just like the trial did

The mpr and pcr rates are low. This isn’t like lung. Also there’s no chemo, which drives path responses.

 

[medgator]

Treat per path indications just like the trial did

The mpr and pcr rates are low. This isn’t like lung. Also there’s no chemo, which drives path responses.

The reps have already visited me to push for this since I refer out for the big surgeries sometimes after seeing pts from local omfs or ent's

From what I understand the trial showed decreased need for CRT vs post op RT alone given the decrease in +margins and/or +ece in the IO arm

 

[TheWallnerus]

Hey!

Our med oncs are starting to push for perioperative Pembro in resectable HNSCC based on this trial.

https://www.nejm.org/doi/full/10.1056/NEJMoa2415434

What are the implications for adjuvant RT?

The trial treated pretty much everyone with at least 60 Gy. I am wondering what am I supposed to in patients achieving major pathological response. Do I still treat the same way?

Have you seen this happening at your institutions?

What would be very cool, and trenchant, would be if we rad oncs started testing doses as low as 10 (don't laugh), 20, or 30 Gy in settings like this. We are after all already toying around with 30 Gy in head/neck cancer anyway. The great undiscovered country in rad onc HNSCC is dose de-escalation. It was predicted, kind of, long ago (Fletcher) that this was the way. Wouldn't it be interesting if A.I. could model the number of remaining clonogens to inform dosing choices after a systemic treatment since we know, as Fletcher knew, "that all epithelial tumors,such as squamous cell carcinomas, adenocarcinomas of the breast and the uterus, and mucoepidermoid, malignant-mixed, and adenoid-cystic carcinomas of the salivary glands, have approximately the same radiosensitivity, other factors being equal."

Treat per path indications just like the trial did

The mpr and pcr rates are low. This isn’t like lung. Also there’s no chemo, which drives path responses.

The degree that chemo drives path responses is very correlated to how well the immunotherapy works though (think: certain rectal cancers, especially). I remember when Eli Glatstein said to me "We are just one drug away from being out of the lymphoma game."

“Six years ago, a medical oncologist told me he had never seen a patient with HER2-positive breast cancer develop local recurrence. I set out to find patients who had recurrence in this setting. Approximately 2,000 patients later, I have not seen one, either."

 

[radonc17]

Hey!

Our med oncs are starting to push for perioperative Pembro in resectable HNSCC based on this trial.

https://www.nejm.org/doi/full/10.1056/NEJMoa2415434

What are the implications for adjuvant RT?

The trial treated pretty much everyone with at least 60 Gy. I am wondering what am I supposed to in patients achieving major pathological response. Do I still treat the same way?

Have you seen this happening at your institutions?

We are doing this per surgeon preference is they feel the upfront surgery is going to be terrible.

12% pts in trial had a delay to surgery so that's something to consider for surgeons

~20% didn't receive adjuvant therapy -> in the Pembro arm half of those were due to progression of disease

If patients made it to the adjuvant phase, they all received radiotherapy so no change in how we are treating on our end.

 

[OTN]

What would be very cool, and trenchant, would be if we rad oncs started testing doses as low as 10 (don't laugh), 20, or 30 Gy in settings like this. We are after all already toying around with 30 Gy in head/neck cancer anyway. The great undiscovered country in rad onc HNSCC is dose de-escalation. It was predicted, kind of, long ago (Fletcher) that this was the way. Wouldn't it be interesting if A.I. could model the number of remaining clonogens to inform dosing choices after a systemic treatment since we know, as Fletcher knew, "that all epithelial tumors,such as squamous cell carcinomas, adenocarcinomas of the breast and the uterus, and mucoepidermoid, malignant-mixed, and adenoid-cystic carcinomas of the salivary glands, have approximately the same radiosensitivity, other factors being equal."

View attachment 409486

The degree that chemo drives path responses is very correlated to how well the immunotherapy works though (think: certain rectal cancers, especially). I remember when Eli Glatstein said to me "We are just one drug away from being out of the lymphoma game."

“Six years ago, a medical oncologist told me he had never seen a patient with HER2-positive breast cancer develop local recurrence. I set out to find patients who had recurrence in this setting. Approximately 2,000 patients later, I have not seen one, either."

Adjuvant 30 Gy H+N trial was negative though with more recurrences compared with standard of care

 

[temujim]

Hey!

Our med oncs are starting to push for perioperative Pembro in resectable HNSCC based on this trial.

https://www.nejm.org/doi/full/10.1056/NEJMoa2415434

What are the implications for adjuvant RT?

The trial treated pretty much everyone with at least 60 Gy. I am wondering what am I supposed to in patients achieving major pathological response. Do I still treat the same way?

Have you seen this happening at your institutions?

A bit of a garbage trial. 10% of the standard of care arm noped out (vs 1% on the experimental arm) -- who knows what happened to them. The company certainly does not. And you see that in the KM curves -- essentially the entire difference occurs early -- if these patients went off and did something non-standard you might expect this result.

They received an indication for CPS >=1, but all the responses were essentially CPS >=10. If you're going to do this regimen, certainly should be restricted to this group, which was around 60% of the total.

PFS was positive, but OS was not by their own statistical testing, even in CPS >= 10. Time will tell on that. But the FDA allowing PFS endpoint to drive indications instead of OS is irritating.

The other concern is toxicity reporting -- frankly unbelievable. And there is a complete absence of surgical related toxicity reporting. Eg roughly 30% of the total population was larynx/hypopharynx and there were no permanent tracheostomies reported. But they do report radiation skin injury!

Anyway another NEJM fail.

 

[ramsesthenice]

A bit of a garbage trial. 10% of the standard of care arm noped out (vs 1% on the experimental arm) -- who knows what happened to them. The company certainly does not. And you see that in the KM curves -- essentially the entire difference occurs early -- if these patients went off and did something non-standard you might expect this result.

They received an indication for CPS >=1, but all the responses were essentially CPS >=10. If you're going to do this regimen, certainly should be restricted to this group, which was around 60% of the total.

PFS was positive, but OS was not by their own statistical testing, even in CPS >= 10. Time will tell on that. But the FDA allowing PFS endpoint to drive indications instead of OS is irritating.

The other concern is toxicity reporting -- frankly unbelievable. And there is a complete absence of surgical related toxicity reporting. Eg roughly 30% of the total population was larynx/hypopharynx and there were no permanent tracheostomies reported. But they do report radiation skin injury!

Anyway another NEJM fail.

Preopanc-2 just came out last week showing no OS for FOLFIRINOX compared to single agent gem and CRT for borderline resectable pancreatic cancer. Also reported no differences in toxicity. Of course, things like neuropathy were not presented (at least in the “it’s all good” abstract.) Was Lancet instead of NEJM, but still, they can all do better.

 

[Palex80]

Thank you, all! I was contemplating on whether or not, I can alter my radiotherapy schedule based on the pathological response.
The trial gave at least 60 Gy, ECOG-ACRIN 3311 was not published by the time they designed the trial. And we now know that we can drop the post-op dose to 50 Gy to patients in the intermediate risk group.

1. If I see major pathological response, could I drop the dose to the contralateral neck (for instance down to 30 Gy) or even ommit it completely?

2. Am I allowed to give only 50 Gy, if they are in the intermediate-risk group post-neoadjuvant Pembro as per ECOG-ACRIN-criteria?

 

[drowsy12]

3311 is for HPV positive patients. These patients are not the keynote 689 population.

follow standard dose recommendations.

 

[temujim]

Thank you, all! I was contemplating on whether or not, I can alter my radiotherapy schedule based on the pathological response.
The trial gave at least 60 Gy, ECOG-ACRIN 3311 was not published by the time they designed the trial. And we now know that we can drop the post-op dose to 50 Gy to patients in the intermediate risk group.

1. If I see major pathological response, could I drop the dose to the contralateral neck (for instance down to 30 Gy) or even ommit it completely?

2. Am I allowed to give only 50 Gy, if they are in the intermediate-risk group post-neoadjuvant Pembro as per ECOG-ACRIN-criteria?

Should emphasize that ECOG 3311 was a feasibility trial, not a non-inferiority trial. Sure things looked good, but so did HN002. Continue to believe 60 Gy is standard of care.

Elective neck I think it's reasonable to go down to 40 Gy. Still think 30 Gy is a bridge too far.

 

[RickyScott]

https://ascopubs.org/doi/10.1200/JCO-24-02194

43 Gy/1.26 fraction without chemotherapy

Head and Neck Cancer: Elective Nodal De-Escalation

A New Era in elective nodal dosing is replacing a 60+ year standard.

protons101.substack.com

 

[TheWallnerus]

I think 60 Gy to an area that was clinically positive and is now path negative after “chemo”therapy (I was reading an old German article about RT for mastitis… it works great!… and they called the antibiotics “chemotherapy”) is a bit much. Based on the HN data of the past decade like RickyScott is linking to, I would decrease dose less than 60. While LR is certainly very bad, every gray we can intelligently spare is a better toxicity profile as we all know. Look back at the old Fletcher subclinical disease ruminations implying ~40 Gy should give about 90-99% LRC in a pCR setting.

 

[OTN]

https://ascopubs.org/doi/10.1200/JCO-24-02194

43 Gy/1.26 fraction without chemotherapy

Head and Neck Cancer: Elective Nodal De-Escalation

A New Era in elective nodal dosing is replacing a 60+ year standard.

protons101.substack.com

Idk, that discussion is pretty wild. I'm not sure how you can see randomized data pointing one way (not even including the recently published Mayo data which was also negative), non-randomized data pointing the other, and go with the non-randomized answer. The failure of dose-escalation in Stage III NSCLC should have been a warning to us all.

 

[Palex80]

3311 is for HPV positive patients. These patients are not the keynote 689 population.

follow standard dose recommendations.

True, the last couple of patients I saw had p16+ disease.
HPV-positive tumors were allowed in Keynote689, but they were a tiny proportion (around 4%).
But there is quite some evidence pointing at immunotherapy working very well for the p16 tumors.

 

[drowsy12]

The use of it is meant for HPV negative resectable patients. Keep a consistent straight forward approach.

The study shows decrease in distant mets. Has been hard to do that in HNSCC. OS is pending.

Immunotherapy in this setting does not give robust local responses, either clinically or pathologically.

 

[Palex80]

The use of it is meant for HPV negative resectable patients.

The trial allowed all tumors, HPV positive and negative. It actually allowed HPV positive oropharyngeal cancer with T4 N0-N2, making the whole "resectable" criterium a bit vague. Technically, a T4 can be resectable, whether or not it's a good idea, is another issue.

Keep a consistent straight forward approach.

Please translate.

 

[drowsy12]

The trial allowed all tumors, HPV positive and negative. It actually allowed HPV positive oropharyngeal cancer with T4 N0-N2, making the whole "resectable" criterium a bit vague. Technically, a T4 can be resectable, whether or not it's a good idea, is another issue.


Please translate.

1) As you said the trial barely included HPV positive patients. In the US, that has resulted in the NCCN guidelines only recommending 689 regimen for HPV negative patients. this is how oncs are interpreting it too. it doesn't really make too much sense to try to escalate therapy in HPV-positive disease anyways.

2) this entire thread is an over complication so I will repeat myself again. it's pretty straight forward. you treat based on path findings, as you do now. positive margins, ECE, etc buys you chemo. and may buy you 66 in common practice. otherwise 60/30.