Nivolumab for Recurrent or Metastatic Head and Neck Squamous Cell Carcinoma: A Retrospective Tertiary Centre’s Real-World Experience

Round 1

Reviewer 1 Report (Previous Reviewer 3)

Brief summary 

After corrections and editing, the paper seems more precise and complete.  

Still some minor corrections should be made 

Specific comments: 

ABSTRACT 

 “Prior to nivolumab, 44% of patients had surgery, 97% radiation, and 100% chemotherapy” please specify here if the prior treatments were used for the primary tumor or the recurrent one  

 
Introduction 

- Line 31-32 please be more accurate in the epidemiology of the disease.  

Material and methods  

- please be more accurate describing the inclusion criteria: age of onset at least 18 years old, performance status of the patients as you reported ECOG score later in the text, hematologic, hepatic and kidney function and so on  

- please specify if you selected patients using also exclusion criteria, as autoimmune disease, certain metastatic localisation, immunodepression and so on 

- have you information of tumor PD-L1 expression? If you have them, pleas add it to the text. CHECKMATE 141 study tried to correlate PDL1 expression with nivolumab efficacy  

Outcomes 

- do you also note the time of the response to nivolumab? 

Author Response

We thank Reviewer 1 for taking the time to assess our article and provide feedback. 

Specific comments:

ABSTRACT 

“Prior to nivolumab, 44% of patients had surgery, 97% radiation, and 100% chemotherapy” please specify here if the prior treatments were used for the primary tumor or the recurrent one  

 Reply: 

We have outlined in the line following after this line. Most (97%) were for primary disease.Introduction 

- Line 31-32 please be more accurate in the epidemiology of the disease.  

 We outlined some specific statistics (highlighted in red) in our introduction but we did not want to overwhelm the readers with too many statistics as the begin to read this piece. 

Material and methods  

• please be more accurate describing the inclusion criteria: age of onset at least 18 years old, performance status of the patients as you reported ECOG score later in the text, hematologic, hepatic and kidney function and so on 

Reply: 

We have included age of onset being at least 18 years old into our methods. Although we do comment on ECOG later on, we did not use it as a basis of selecting our cohort as we wanted our sample to be as realistic of a population as possible. The same applies for hepatic and kidney function.

• please specify if you selected patients using also exclusion criteria, as autoimmune disease, certain metastatic localisation, immunodepression and so on

Reply: We wanted our population to be as applicable to a real-world population of patients receiving Nivolumab as possible. Therefore, all of the exclusion criteria listed is what we used to create our study population.

• have you information of tumor PD-L1 expression? If you have them, pleas add it to the text. CHECKMATE 141 study tried to correlate PDL1 expression with nivolumab efficacy 

Reply: Unfortunately, our Canadian institutions had not been collecting PD-L1 expression information during the time period of our study so we are not able to comment on this.

Outcomes 

• do you also note the time of the response to nivolumab?

Reply:

We were not able to provide an accurate time of response to Nivolumab data due to the nature of follow up visit intervals. 

Reviewer 2 Report (Previous Reviewer 4)

Thank you for your improvements

Nothing to add

Author Response

We thank Reviewer 1 for taking the time to review our article. 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The authors present their experience using an immune checkpoint inhibitor for R/M HNSCC following the CHECKMATE trial. Due to differences in clinicodemographic variables, small number of patients, and no standardized radiographic evaluation of objective response, their results are challenging to interpret. There is value in having more data on ICI responses in the literature and a wider group of patients. I think with major revisions this could be of interest to the head and neck oncology community. The most important revision is establishing a systematic way of evaluating objective response rate – ideally this would involve partnering with a radiologist to go over the CT scans and apply RECIST criteria or some other type of volumetric assessment of the tumor and metastatic sites.

Abstract

Overall a nicely worded and succinct abstract. Only comment is that the last line claiming "Nivolumab appears to have decreased real-world survival rates…" may be too strong of a conclusion based on a limited (n = 34) number of patients. I would consider diluting this down statement to reflect the heterogeneity of the R/M HNSCC cohort overall.

Introduction

Nice, concise summary of the relevant literature.

Methods

Lines 54-58: Wording is extremely confusing on inclusion criteria. Were all these patients recurrences within 6 months of treatment? It also sounds like an exclusion criteria was not having another primary cancer. Divide inclusion and exclusion criteria into separate sentences.

Lines 54-58: Explain why recurrences within 6 months were chosen as an inclusion criteria. Why not one year or some other date?

Lines 64-66: How was clinical response determined on imaging if not using a radiologist or RECIST criteria? Were the authors just reading imaging reports or actually looking at the scans? Need to be more specific since this is the major criteria to judge objective responses and could bias the entire paper.

Line 71: Please state whether ORR specifically refers to radiographic response.

Results

Line 95/Table 1: Are the stage groups the initial stage or recurrence? It would be helpful to include initial stage groups so we can have a sense of the severity of the initial tumor.

Line 156: Why are you assuming a treatment response without objective data? These 5 patients should be excluded (at least from Figure 2 for measuring PFS) if they do not have post-treatment imaging. It may be OK to include them in the overall survival figure.

Discussion

Lines 187-188: You previously state 97% had radiation and 44% had surgery but the stat is reverse in this line. Please correct the typo.

Lines 201-203: Are there any papers that evaluate RECIST criteria vs radiology reports in outcomes?

Add a separate discussion paragraph on adverse event rates and risks/benefits of treatment that has limited therapeutic efficacy.

Author Response

Specific comments:

Abstract:

Overall a nicely worded and succinct abstract. Only comment is that the last line claiming "Nivolumab appears to have decreased real-world survival rates…" may be too strong of a conclusion based on a limited (n = 34) number of patients. I would consider diluting this down statement to reflect the heterogeneity of the R/M HNSCC cohort overall.

Response: The last line was changed to reflect our single-centre canadian population: Nivolumab appears to have decreased survival rates in our single-centre Canadian population compared to CHECKMATE-141 and a manageable adverse event profile for R/M HNSCC.

Introduction

Nice, concise summary of the relevant literature.

Methods

Lines 54-58: Wording is extremely confusing on inclusion criteria. Were all these patients recurrences within 6 months of treatment? It also sounds like an exclusion criteria was not having another primary cancer. Divide inclusion and exclusion criteria into separate sentences.

Response: Inclusion and exclusion criteria were clarified: “Key eligibility criteria included receiving nivolumab for R/M HNSCC, having HNSCC confirmed on initial histology, experienced recurrent or progressive cancer within 6 months of their last dose of platinum-containing chemotherapy for primary or recurrent HNSCC. Individuals were excluded if they received systemic treatment for any other primary cancer within 6 months of their first dose of nivolumab.”

Lines 54-58: Explain why recurrences within 6 months were chosen as an inclusion criteria. Why not one year or some other date?

Response: This criterion was used to match the Canadian nivolumab label, which is that nivolumab is indicated for patients with relapse within 6 months of completing definitive chemo-radiation, or after progressing on platinum based chemotherapy for r/m HNSCC.

Lines 64-66: How was clinical response determined on imaging if not using a radiologist or RECIST criteria? Were the authors just reading imaging reports or actually looking at the scans? Need to be more specific since this is the major criteria to judge objective responses and could bias the entire paper.

Response: The response was abstracted from the radiologists clinical report, which were not reported according to RECIST criteria. We acknowledge this limitation in the paper.

Line 71: Please state whether ORR specifically refers to radiographic response.

Response: ORR refers to radiographic response and we changed our outcomes section to reflect this definition.

Results

Line 95/Table 1: Are the stage groups the initial stage or recurrence? It would be helpful to include initial stage groups so we can have a sense of the severity of the initial tumor.

Response: We collected the stage group at initial diagnosis of SCCHN. This has been clarified in the table.

Line 156: Why are you assuming a treatment response without objective data? These 5 patients should be excluded (at least from Figure 2 for measuring PFS) if they do not have post-treatment imaging. It may be OK to include them in the overall survival figure.

Response: Thank you for this comment. We have removed the 5 patients from the Kaplan-Meier analysis for PFS. These changes are reflected on lines 162-163, 174-179, and the updated Kaplan-Meier diagram on page 6.

Discussion

Lines 187-188: You previously state 97% had radiation and 44% had surgery but the stat is reversed in this line. Please correct the typo.

Response: The mixup between percentages for radiation and surgery have been fixed. Thank you for catching this mistake.

Lines 201-203: Are there any papers that evaluate RECIST criteria vs radiology reports in outcomes?

Response: After doing a search through the literature, we were not able to find any papers that evaluate RECIST vs radiology reports in outcomes

Add a separate discussion paragraph on adverse event rates and risks/benefits of treatment that has limited therapeutic efficacy.

Response: We have added a paragraph that explores the risks/benefits of Nivolumab therapy in our discussion.

Reviewer 2 Report

Dear authors

I have read your manuscript with great interest and complement you on the work performed.

Herewith I provide you my major remarks:

- Why did you not include patients treated in 2021 and 2022? Although you state that pembrolizumab is now the preferred treatment option, you assume there will be still some patients who have been treated with nivolumab in the past 2 years. I therefore propose to also include these patients into the dataset.

- You acknowledge that you have only a little number of patients in the dataset. Is it therefore not possible to also include patients from other Canadian centers? This would greatly increase the total number of patients in your study and results in a more representative population.

- In the discussion, you only compare your real-life data to the initial data published from the Checkmate-141 study. I propose to also compare your data to more recently published clinical trials using nivolumab in R/M SCCHN. Hereby, I refer to

- Checkmate-141 (long-term follow up), Gillison et al. Oncologist 2022, PMID 35641218

- Checkmate-714 (nivolumab comparator arm), Harrington et al. JAMA Oncol 2023, PMID 37022706

My minor remarks can be found in the document attached.

Comments for author File: Comments.pdf

Author Response

Specific Major comments:

- Why did you not include patients treated in 2021 and 2022? Although you state that pembrolizumab is now the preferred treatment option, you assume there will be still some patients who have been treated with nivolumab in the past 2 years. I therefore propose to also include these patients into the dataset.

Response: We did not include the patients treated in 2021 and 2022 as our current database did not extend to include those patients. We felt that the data from 2017-2020 would provide a good picture of the drug’s use during the years when its use was most prominent to ensure our demographic analysis would be more unified.

- You acknowledge that you have only a little number of patients in the dataset. Is it therefore not possible to also include patients from other Canadian centers? This would greatly increase the total number of patients in your study and results in a more representative population.

Response: It would have been very interesting to include the date from other Cancer centres. We can perhaps do this for our next project. Unfortunately, in Canada, our institutional data is quite fragmented and not easily accessible for research purposes. We do feel that the data from our centre, although smaller in numbers, is of value to bring light to Nivolumab’s use in a large Canadian cancer centre.

- In the discussion, you only compare your real-life data to the initial data published from the Checkmate-141 study. I propose to also compare your data to more recently published clinical trials using nivolumab in R/M SCCHN. Hereby, I refer to

- Checkmate-141 (long-term follow up), Gillison et al. Oncologist 2022, PMID 35641218

- Checkmate-714 (nivolumab comparator arm), Harrington et al. JAMA Oncol 2023, PMID 37022706

Response: Thank you for providing us with these two interesting studies. We have included two other papers published recently (one from England and one from Japan) that investigates the use of Nivolumab in a real world setting. The timelines of these studies are similar to ours and we can therefore compare our overall survival rates.

Minor Suggestions: All of Reviewer 2’s minor edits in an attached document have been addressed in the edited manuscript.

Reviewer 3 Report

Brief summary

The paper focuses the attention on an interesting topic, as new therapeutical strategies are always needed to improve patients outcomes in cases of recurrence or progression of disease.
This is particularly true in head and neck cancer, as is it an aggressive disease and lot of patients experience relapses or metastases during disease.

This is a clear paper, with a good structure.

The conclusions are interesting and contribute to originally review the literature.

No ethical problems are found in this study.

However, some criticisms are present in the paper.

Specific comments:

ABSTRACT

- “The median 17 patient age at nivolumab initiation was 61, 24% were female, and 62% were current or former smokers” - I suggest being more schematic with these information

- “Prior to nivolumab, 44% of patients had surgery, 97% radiation, and 100% chemotherapy” please specify here if the prior treatments were used for the primary tumor or the recurrent one

- “18% of patients experienced an adverse immune event, the most common of which were pneumonitis (9%) and endocrine (9%)” Please specify the number of patients who experienced adverse events, as right after you write “7/8 of the immune adverse effect” but the sentence is not clear Introduction

- Line 31-32 please be more accurate in the epidemiology of the disease.

- Line 32-34: please add recurrence or progression of disease scores and percentage of patients.

Material and methods

- please be more accurate describing the inclusion criteria: age of onset at least 18 years old, performance status of the patients as you reported ECOG score later in the text, hematologic, hepatic and kidney function and so on

- please specify if you selected patients using also exclusion criteria, as autoimmune disease, certain metastatic localisation, immunodepression and so on

- I suggest to specify, if it is possible, Nivolumab therapeutical scheme of administration as in CHECKMATE 141 study “ Nivolumab was administered at a dose of 3 mg per kilogram of body weight every 2 weeks.” (Ferris RL, Blumenschein G Jr, Fayette J, Guigay J, Colevas AD, Licitra L, Harrington K, Kasper S, Vokes EE, Even C, Worden F, Saba NF, Iglesias Docampo LC, Haddad R, Rordorf T, Kiyota N, Tahara M, Monga M, Lynch M, Geese WJ, Kopit J, Shaw JW, Gillison ML. Nivolumab for Recurrent Squamous-Cell Carcinoma of the Head and Neck. N Engl J Med. 2016 Nov 10;375(19):1856-1867. doi: 10.1056/NEJMoa1602252. Epub 2016 Oct 8. PMID: 27718784; PMCID: PMC5564292.)

- have you information of tumor PD-L1 expression? If you have them, pleas add it to the text. CHECKMATE 141 study tried to correlate PDL1 expression with nivolumab efficacy

Outcomes

- do you also note the time of the response to nivolumab?

- line 95 “from 2018 to 2020”

- line 106-108: please report the number of patients with the percentage of metastatic events as you reported in the table

- line 111-118 line of treatment: this part is a little confusing, please write it again in a more schematic way; always put together the number of patient and the percentage

Discussion

- line 186: “an advanced cancer stage (at diagnosis?)”. I suggest deleting “at diagnosis?”

only minor revisions needed

Author Response

ABSTRACT

- “The median 17 patient age at nivolumab initiation was 61, 24% were female, and 62% were current or former smokers” - I suggest being more schematic with these information

Response: Although these visualization techniques may be helpful, our sample size is very small which would make the visualizations less informative and for this analysis, the most important visualizations are for the survival analysis which we have provided (Kaplan-Meier analysis).

- “Prior to nivolumab, 44% of patients had surgery, 97% radiation, and 100% chemotherapy” please specify here if the prior treatments were used for the primary tumor or the recurrent one.

Response: We added the following lines after this to specify what prior treatments were used for: Most (97%) were for primary disease.

- “18% of patients experienced an adverse immune event, the most common of which were pneumonitis (9%) and endocrine (9%)” Please specify the number of patients who experienced adverse events, as right after you write “7/8 of the immune adverse effect” but the sentence is not clear

Response: We have made our reporting of adverse events more consistent in this area by using the number of patients to describe all adverse events.

Introduction

- Line 31-32 please be more accurate in the epidemiology of the disease.

Response: We have added specific epidemiological figures to our introduction.

- Line 32-34: please add recurrence or progression of disease scores and percentage of patients.

Response: We have added rates of recurrence figures to our introduction.

Material and methods

- please be more accurate describing the inclusion criteria: age of onset at least 18 years old, performance status of the patients as you reported ECOG score later in the text, hematologic, hepatic and kidney function and so on

Response: We clarified our inclusion and exclusion criteria. We chose to have a limited list of inclusion criteria to represent the real-world use of Nivolumab.

- please specify if you selected patients using also exclusion criteria, as autoimmune disease, certain metastatic localisation, immunodepression and so on

Response: We clarified our exclusion criteria.

- I suggest to specify, if it is possible, Nivolumab therapeutical scheme of administration as in CHECKMATE 141 study “ Nivolumab was administered at a dose of 3 mg per kilogram of body weight every 2 weeks.” (Ferris RL, Blumenschein G Jr, Fayette J, Guigay J, Colevas AD, Licitra L, Harrington K, Kasper S, Vokes EE, Even C, Worden F, Saba NF, Iglesias Docampo LC, Haddad R, Rordorf T, Kiyota N, Tahara M, Monga M, Lynch M, Geese WJ, Kopit J, Shaw JW, Gillison ML. Nivolumab for Recurrent Squamous-Cell Carcinoma of the Head and Neck. N Engl J Med. 2016 Nov 10;375(19):1856-1867. doi: 10.1056/NEJMoa1602252. Epub 2016 Oct 8. PMID: 27718784; PMCID: PMC5564292.)

Response: Nivolumab was administered at 6mg/kg (up to a maximum of 480mg) IV every 4 weeks. This has been reflected in lines 97-98.

- have you information of tumor PD-L1 expression? If you have them, pleas add it to the text. CHECKMATE 141 study tried to correlate PDL1 expression with nivolumab efficacy

Response: PD-L1 testing was not standard of care during the time frame of this study. Nivolumab could be prescribed without knowing PD-L1 status, and therefore none of the patients had this testing done.

Outcomes

- do you also note the time of the response to nivolumab?

Response: We did not collect duration of response data

- line 95 “from 2018 to 2020”

Response: This has been fixed to reflect the timeline.

- line 106-108: please report the number of patients with the percentage of metastatic events as you reported in the table

Response: We had reported the number of patients with the percentage of metastatic events.

- line 111-118 line of treatment: this part is a little confusing, please write it again in a more schematic way; always put together the number of patient and the percentage

Response: We have added the number of patients and the percentage to each of the points in this paragraph. We have also clarified the data presented by ensuring the temporality to Nivolumab administration is specified.

Discussion

- line 186: “an advanced cancer stage (at diagnosis?)”. I suggest deleting “at diagnosis?”

Response: We have removed this. Thank you for catching this error.

Reviewer 4 Report

Congratulations to the authors for the work done. Well-structured.

I have a question regarding the inclusion criteria and study participants. In the inclusion criteria, it is stated that there should be a 6-month period since the last administration of systemic chemotherapy treatment. How is it possible that patients with a shorter time than this period are included? Line 97-98: "The median follow-up time was 9.4 (interquartile range [IQR], 4.7-14.3) months from the date of cancer diagnosis."

In the 11 cases of oropharynx, was the HPV status not evaluated?

Indeed, conducting subgroup analyses focusing specifically on oropharyngeal and oral cavity cases, with corresponding Kaplan-Meier survival curves, could add significant value to the study. This would allow for a more targeted examination of the outcomes in these particular subgroups and help identify any potential differences in survival rates between them.

Additionally, evaluating the quality of life or the severity of adverse reactions would be an interesting aspect to explore. Understanding how adverse reactions impact patients' well-being and survival could provide valuable insights for healthcare professionals and researchers. It would be important to investigate whether patients who experienced such adverse reactions had lower survival rates or different treatment outcomes compared to those who did not experience them. Including this information in the study would provide a more comprehensive understanding of the patients' experiences and outcomes.

I believe it is a feasible analysis given the small sample size.

I think it is advisable to delve deeper into the limitations of the study.

Author Response

I have a question regarding the inclusion criteria and study participants. In the inclusion criteria, it is stated that there should be a 6-month period since the last administration of systemic chemotherapy treatment. How is it possible that patients with a shorter time than this period are included? Line 97-98: "The median follow-up time was 9.4 (interquartile range [IQR], 4.7-14.3) months from the date of cancer diagnosis."

Response: Thank you for bringing up this point. We have clarified this line to reflect the “date of cancer recurrence diagnosis”.

In the 11 cases of oropharynx, was the HPV status not evaluated?

Indeed, conducting subgroup analyses focusing specifically on oropharyngeal and oral cavity cases, with corresponding Kaplan-Meier survival curves, could add significant value to the study. This would allow for a more targeted examination of the outcomes in these particular subgroups and help identify any potential differences in survival rates between them.

Response: We attempted to collect HPV status on our patients in hopes of running subgroup analysis. Unfortunately, HPV status was not reported for many of our patients and this could not be done.

Additionally, evaluating the quality of life or the severity of adverse reactions would be an interesting aspect to explore. Understanding how adverse reactions impact patients' well-being and survival could provide valuable insights for healthcare professionals and researchers. It would be important to investigate whether patients who experienced such adverse reactions had lower survival rates or different treatment outcomes compared to those who did not experience them. Including this information in the study would provide a more comprehensive understanding of the patients' experiences and outcomes.

I believe it is a feasible analysis given the small sample size.

I think it is advisable to delve deeper into the limitations of the study.

Response: Thank you for this thoughtful comment. Our research group has been working on incorporating more qualitative analysis into our larger areas of study to bring the patient voice to literature. We feel that this would be a great project to take on in the future as the analysis of qualitative data is quite different to what we have done here. We will definitely consider doing this as we also are curious around experiences with adverse reactions.