Exploring the Frequency and Risk Factors of Hyperprogressive Disease in Patients with Advanced Melanoma Treated with Immune Checkpoint Inhibitors

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

Hyperprogressive disease (HPD), a paradoxically rapid disease progression in advanced cancer patients following treatment, remains a somewhat polarising concept due to the lack of clearly defined criteria. In this single-centre retrospective study, Acar et al analyzed a cohort of 158 advanced melanoma patients treated with immunotherapy (anti-PD1 +/- anti-CTLA4), for HPD occurrence and risk factors. HPD defined according to previously published criteria (Lo Russo CCR 2018), was found in 24 patients (15.2%). Multivariant analysis of clinicopathological and laboratory variables identified ECOG performance score, elevated LDH, presence of liver metastasis, multiple metastatic sites and low eosinophil counts, as factors significantly associated with HPD. In addition, performance of three Phase I trials prognostic scores (MDA-ICI, RMH and GRIm) was assessed for predicting HPD in this patient cohort, with RMH and MDA-ICI scores (variably including some of the above parameters) also performing well.

 

Identifying patients with advanced melanoma who are at risk of rapid disease progression, is critical for optimal therapy selection. Several reports have previously addressed HPD occurrence and risk factors in melanoma patients receiving immune checkpoint inhibitors, as well as other treatments. This study is an important addition to the field. However, improved data transparency is critical.

For the 24 patients with HPD, could the authors provide information contained in Table 1, for each patient individually, in either table or figure format. Any connection between the individual BRAF status, type and number of organ metastases, type of ICI, LDH, ECOG, lab values etc, is currently lost in statistical comparisons. Please also include information on ICI discontinuation due to irAEs/toxicity, and on steroid use.

Just to note, Table S1 may contain some of this information, it is mentioned in the text but was not available for review.

 

Line 158. Please clarify, “For the MDA-ICI and RMH scores, multivariate analysis was performed by modelling without including the parameters used in these scores”.

 

Lack of predictive value for SII, PIV, SIRI and HALP scores is intriguing and may reflect distinct inflammatory changes in HPD – please add to the discussion.

 

Minor.

Figure 1, please correct the x-axis values.

Table 2 title. Please move to next page.

Table 1, “Previous ICI”- CTLA-4 not CLTA-4; GRIm not Grim score

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is an interesting subject matter, which has received relatively little attention to date in melanoma. 

My major concern is comparing hyperprogression in a cohort which received anti-PD1 monotherapy and one which received immune combination (anti-PD1 and anti-CTLA4 therapy). The groups are likely to be clinically different, both in terms of tumor burden and prognosis, which probably led to the decision to administer these treatments.

How can the authors exclude this as an important confounding variable?

I think that the introduction could be more comprehensive and the rationale for using a plethora of scores should be justified and explained.

The clinical consequences of the work need to be carefully crystallized. 

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for carefully revising the manuscript.