Cancer Risk in Men with HIV in Japan: An 18-Year Single-Center Cohort Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. In Supplementary data file (for both Table 1 & 2), it seems that few data rows are non-significant (P value >0.05). Why authors presented those data?
2. In summery authors stated that “We found that the risk of AIDS-defining cancers has decreased dramatically. The overall risk of other cancers has also decreased to levels similar to those of the general population. However, the risk for some specific cancers, particularly anal cancer and oral/pharyngeal”
   Which cancers are AIDS-defining cancers and what do you understand by other cancers. This section should be written clearly.
3. In next line authors told that “However, the risk for some specific cancers, particularly anal cancer and oral/pharyngeal cancers, remains persistently high.” Do these cancers fall under ‘AIDS-defining cancers’ or ‘other cancers’?
4. In methods section (line 119-120), what is the criteria for defining ADM and NADM. Authors stated NADMs (all others). What does this all other means, as there is lots of cancer types of presents.
5. Authors finally included 3,793 men with HIV for the study, but they identify only 288 malignant cases. Please specify the stage of cancers (Stage 1/2/3). 

Author Response

Comment 1: In Supplementary data file (for both Table 1 & 2), it seems that few data rows are non-significant (P value >0.05). Why authors presented those data?

Response 1: We included both significant and non-significant results in the Supplementary Tables to provide a complete and transparent description of cancer incidence patterns in this cohort. Restricting the Supplementary Tables to statistically significant findings alone could create an impression of selective reporting and might prevent readers from interpreting the overall pattern of site-specific risks.

 

Comment 2: In summary authors stated that "We found that the risk of AIDS-defining cancers has decreased dramatically. The overall risk of other cancers has also decreased to levels similar to those of the general population. However, the risk for some specific cancers, particularly anal cancer and oral/pharyngeal cancers..." Which cancers are AIDS-defining cancers and what do you understand by other cancers. This section should be written clearly.

Response 2: We have now defined AIDS-defining cancers as Kaposi’s sarcoma and AIDS-related lymphoma and “other cancers” as non-AIDS-defining malignancies (i.e., all malignancies other than the specified AIDS-defining cancers) in the Simple Summary (lines 15–18).

 

Comment 3: In next line authors told that "However, the risk for some specific cancers, particularly anal cancer and oral/pharyngeal cancers, remains persistently high." Do these cancers fall under AIDS-defining cancers or other cancers?

Response 3: Anal cancer and oral/pharyngeal cancers were classified as non-AIDS-defining malignancies (NADMs). We have revised the Simple Summary to state this explicitly (lines 19–20).

 

Comment 4: In methods section, what is the criteria for defining ADM and NADM. Authors stated NADMs (all others). What does this all other means, as there is lots of cancer types of presents.

Response 4: ADMs were defined as Kaposi’s sarcoma and AIDS-related non-Hodgkin lymphoma, whereas NADMs were defined as all malignant tumors other than the specified ADMs. For SIR estimation of incident malignancies, cancers diagnosed on or before baseline were treated as prevalent and excluded from incident-case counting. These details have been clarified in the Methods (lines 132–143).

 

Comment 5: Authors finally included 3,793 men with HIV for the study, but they identify only 288 malignant cases. Please specify the stage of cancers (Stage 1/2/3).

Response 5: Tumor stage at diagnosis was not consistently available in our medical records for all cases; therefore, stage-specific analyses could not be performed. We have acknowledged this limitation in the Discussion (lines 437–440).

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review the paper ID: cancers-4075534.

Comments:

Section Introduction:
Describe medical care for HIV persons in more detail, including the year of implementation of ART in this clinic and in Japan, whether it is paid or not, etc.

Methods section:
In the subsection Study design:
- Explain the number 4,261 (on Lines 83-87) in relation to the number 4,196 (on Figure 1).
In the Study population and period subsection specify:
- Was Dg HIV established in that period for the first time?
- Was Dg HIV established before the observed period?
- Did and how many HIV subjects already at the beginning of the observed period and/or at the beginning of inclusion in the study already had Dg of a malignant tumor?
- Why was 2007 the year set for the start of follow-up?
- Why is there an unequal number of follow-up years in this study in 4 different length periods by time period from 2007 to 2024?

Results section:
- Line 147: Correct the title of this figure as Figure S1.
- Table 2: Transfer values ​​listed under Table (`Note`) to Table 2.
- In the Results section, the result for rectal cancer is not listed, that is, SIR (95% CI) = 0.38 (0.10–0.97), on Table S1. Correct this, so that this result is stated in the text describing Table S1.
- Lines 217-224: Check and match the title of this figure with the labels for cancer sites on the Figure itself.
- Check if the results for the SIRs for the variable `Non-AIDS-defining malignancies` shown in Table S1 correspond with the results for the same variable shown in Figure 1 (SIRs for `Non-AIDS-defining malignancies`). Correct this.
  
Discussion Section:
- Nowhere in the entire paper is the possible impact of the COVID-19 pandemic discussed, which is entirely covered by the study time period from 2007 to 2024.
Several references can be linked to the results of the current study:
Nishiura H, Fujiwara S, Imamura A, Shirasaka T. Regional variations in HIV diagnosis in Japan before and during the COVID-19 pandemic. Infect Dis Model. 2024;10(1):40-49. doi: 10.1016/j.idm.2024.08.004.

Ejima K, Koizumi Y, Yamamoto N, Rosenberg M, Ludema C, Bento AI, Yoneoka D, Ichikawa S, Mizushima D, Iwami S. HIV Testing by Public Health Centers and Municipalities and New HIV Cases During the COVID-19 Pandemic in Japan. J Acquir Immune Defic Syndr. 2021;87(2):e182-e187. doi: 10.1097/QAI.0000000000002660.

Mitsuya H. Fight against COVID-19 but avoid disruption of services for other communicable diseases (CDs) and noncommunicable diseases (NCDs). Glob Health Med. 2020;2(6):343-345. doi: 10.35772/ghm.2020.01111.

- Lines 274-282: Specifically explain and discuss the finding that the SIR (95%CI) for `Non-AIDS-defining malignancies` in the last period (2021-2024) was 0.69 (0.49–0.95).

- Lines 299-304: In the Discussion section, it is not enough to comment only on the insignificant trend for liver cancer, but to specifically discuss the finding of this study that the following was recorded for liver cancer, for the entire period: SIR (95% CI) = 1.91 (0.99–3.33), on Table S1. Specify the discussion of this result to the circumstances in Japan, citing appropriate references.

- Lines 314-318: Discuss in more detail the finding for rectal cancer, that is, SIR (95% CI) = 0.38 (0.10–0.97), for the entire period. Specify the discussion of this result to the circumstances in Japan.
Note: The explanation `This may reflect a surveillance bias as a result of frequent medical contact among PWH.` is not satisfactory, because such an explanation could be applied to SIRs for all cancer sites. 

 

Author Response

Comment 1: Describe medical care for HIV persons in more detail, including the year of implementation of ART in this clinic and in Japan, whether it is paid or not, etc.

Response 1: We have added Japan-specific context to the Introduction, including the timing of widespread ART availability (late 1990s) and national financial support system that keeps out-of-pocket ART costs minimal, supporting long-term adherence and viral suppression (lines 50–55).

 

Comment 2: Explain the number 4,261 in relation to the number 4,196.

Response 2: We apologize for the inconsistency. The correct cumulative number is 4,196, and we have ensured that 4,196 is consistently mentioned in the manuscript.

 

Comment 3: Was Dg HIV established in that period for the first time? Was Dg HIV established before the observed period?

Response 3: Our cohort includes both (i) individuals newly diagnosed with HIV at our hospital and (ii) individuals diagnosed before the study period and/or at other institutions who subsequently received care at our hospital during the study period. We clarified this in the Methods, including that follow-up begins at the first visit on/after January 1, 2007 (baseline), regardless of when or where HIV was diagnosed (lines 94–109).

 

Comment 4: Did and how many HIV subjects already at the beginning of the observed period and/or at the beginning of inclusion in the study already had Dg of a malignant tumor?

Response 4: Participants were not excluded solely due to a history of malignancy before baseline. However, for SIR estimation of incident malignancies, cancers diagnosed on or before baseline were treated as prevalent and were excluded from incident-case counting (i.e., only malignancies newly diagnosed after baseline contributed to observed incident events). We clarified this approach in the Methods (lines 138–143).

We do not have data on cancers diagnosed before inclusion in the study.

 

Comment 5: Why was 2007 the year set for the start of follow-up?

Response 5: The study period (January 1, 2007 to December 31, 2024) was selected to coincide with the full implementation of the electronic medical record system at our institution, ensuring consistent data capture. This information has been clarified in the Methods (lines 94–96).

 

Comment 6: Why is there an unequal number of follow-up years in this study in 4 different length periods?

Response 6: The four calendar periods were chosen primarily for statistical convenience to ensure sufficient person-years in each period for stable estimation of SIRs and trend testing. We have now explained this in the Statistical Analysis (lines 146–150).

 

Comment 7: Correct the title of this figure as Figure S1.

Response 7: The flow chart has now been referenced as Figure S1 in the Results (line 167).

 

Comment 8: Table 2: Transfer values listed under Table (Note) to Table 2.

Response 8: We have incorporated the note content into Table 2 by listing the relevant cancer types explicitly within the table structure.

 

Comment 9: In the Results section, the result for rectal cancer is not listed (SIR 0.38 (0.10–0.97) on Table S1). Correct this so that this result is stated in the text describing Table S1.

Response 9: We have added the rectal cancer SIR (0.38, 95% CI: 0.10–0.97) to the Results section (lines 228-229) within the summary of site-specific SIRs, consistent with Table S1. We have also expanded the Discussion to interpret this finding in the context of prior large cohort studies (lines 354–365).

 

Comment 10: Check and match the title of this figure with the labels for cancer sites on the Figure itself.

Response 10: Thank you for pointing out the discrepancy between the cancer sites shown in the figure and those reported in the figure legend. We have corrected the Legend of Figure 1C so that the sites reported in the legend correspond to those shown in the figure. In addition, we have added the locations of the individual figures for each site to the legend of Figure 1 (lines 203–212).

 

Comment 11: Check if the results for the SIRs for Non-AIDS-defining malignancies shown in Table S1 correspond with those shown in Figure 1. Correct this.

Response 11: We verified that the NADM SIR values shown in Figure 1 correspond to those shown in Table S1, and no corrections were required.

Comment 12: Nowhere in the entire paper is the possible impact of the COVID-19 pandemic discussed. Several references can be linked to the results of the current study ...

Response 12: We acknowledge that the study period includes the COVID-19 pandemic. However, in our SIR framework, expected counts for 2021–2024 were calculated using population incidence rates from the reference year 2021. Therefore, pandemic-related changes in diagnosis and reporting might have influenced both observed and expected counts, and we believe that this might have limited their net impact on the SIR.

Nonetheless, we recognize that disruptions in healthcare access and diagnostic practices during the pandemic could still have affected cancer ascertainment, particularly for procedures requiring specialist visits, and we have interpreted the most recent-period estimates with this context in mind.

 

Comment 13: Specifically explain and discuss the finding that the NADM SIR in 2021–2024 was 0.69 (0.49–0.95).

Response 13: We have clarified our interpretation of the sub-unity NADM SIR in the Results and Discussion by emphasizing that overall estimates conceal heterogeneity by cancer type and age group (lines 301–303). In our cohort, elevated risks persisted for specific virus-related NADMs (e.g., anal and oral/pharyngeal cancers), whereas several common cancers were comparable to, or lower than, those in the general population, contributing to the overall SIR in the most recent period.

 

Comment 14: Discuss liver cancer (SIR 1.91 (0.99–3.33)) in the circumstances in Japan with references.

Response 14: We substantially expanded the Discussion of liver cancer in the context of Japan (lines 327–342). In this our cohort, hepatocellular carcinoma was more frequently observed in earlier periods among patients with hemophilia who acquired HIV through contaminated blood products and were often coinfected with HCV. While HCV-related hepatocellular carcinoma has been reported as one of the most significant problems in this population in Japan, the contribution of HCV-related hepatocellular carcinoma has declined over time with the availability of direct-acting antiviral therapy. As such, HBV coinfection appeared to contribute minimally to hepatocellular carcinoma development in this cohort.

 

Comment 15: Discuss rectal cancer (SIR 0.38 (0.10–0.97)) in more detail; surveillance bias alone is not satisfactory.

Response 15: We revised and expanded the Discussion of the low rectal cancer SIR by citing evidence from large cohort studies that an inverse association has been observed and persists across tumor stages, suggesting that differential screening alone cannot fully explain the finding. Biologically plausible mechanisms, including HIV-associated disruption of gut mucosal integrity and alterations in gut microbiota composition, have been discussed while emphasizing that these explanations remain speculative and require further investigation (lines 354–365).

Reviewer 3 Report

Comments and Suggestions for Authors

Congratulations to the authors on their work. I find it a very interesting study, both in its approach and conclusions.

The title describes the study and one of its main limitations: its single-center design. I might have written a title like: “Cancer Risk in a Cohort of Men with HIV in Japan: Eighteen Years of Follow-Up.”

The introduction is quite good. It clearly shows the gap they intend to investigate and concludes with a proper formalization of the objectives.

Regarding the materials and methods, I find all the described steps correct. I was surprised by the exclusion of women from the study, but this is justified and mentioned in the study's limitations.

The results are accurate and interesting, as is the discussion, which references other articles.

The bibliographic references are appropriate and up-to-date for this study. I believe it can be published without revisions.

Author Response

Comment 1: Suggested an alternative title and provided an overall positive evaluation.

Response 1: We sincerely thank the reviewer for the positive evaluation and the alternative title suggestion. However, after consideration, we decided to retain the original title because it clearly communicates the study design and single-center setting, which is important for transparency regarding generalizability.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised manuscript may be accepted for publication.

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the opportunity to re-review the manuscript ID: cancers-4075534.

I thank the authors for their efforts in revising this paper.

However, a disingenuous/incorrect answer remained, as follows:

Comment 11: Check if the results for the SIRs for Non-AIDS-defining malignancies shown in Table S1 correspond with those shown in Figure 1. Correct this.
Response 11: We verified that the NADM SIR values ​​shown in Figure 1 correspond to those shown in Table S1, and no corrections were required.

NOTE: THE AUTHORS DID NOT REPLY TO THE COMMENT HONESTLY and CORRECTLY.
Namely, the authors saw the mistake that was indicated by my comment. However, they stated in their response that they checked the NADM SIRs values ​​in Table S1 and Figure 1, and stated that no corrections were needed. 
Namely, the NADM SIRs values ​​presented in Table S1 and Figure 1 must be identical, that is, they must correspond.        

Unfortunately, the authors did not answer honestly, because in the revised version of this paper they entered the modified Figure 1 (so that it matches Table S1) for the variable `Non-AIDS-defining malignancies`.  

The change is evident in Figure 1 in the revised version of this manuscript, although this change was not highlighted in the revised version of this paper, and although the authors denied that those corrections were needed in their response to comments.