The transcriptional landscape analysis of basal cell carcinomas reveals novel signalling pathways and actionable targets

Basal cell carcinoma (BCC) is the most common skin cancer and human malignancy. By analyzing BCC RNA sequencing data according to clinically important features, we identified novel differentially regulated genes and new targetable pathways. Several biomarkers were validated in patient-derived BCC samples.

--An editable version of the final text (.DOC or .DOCX) is needed for copyediting (no PDFs).
--High-resolution figure, supplementary figure and video files uploaded as individual files: See our detailed guidelines for preparing your production-ready images, http://www.life-sciencealliance.org/authors --Summary blurb (enter in submission system): A short text summarizing in a single sentence the study (max. 200 characters including spaces). This text is used in conjunction with the titles of papers, hence should be informative and complementary to the title and running title. It should describe the context and significance of the findings for a general readership; it should be written in the present tense and refer to the work in the third person. Author names should not be mentioned.

B. MANUSCRIPT ORGANIZATION AND FORMATTING:
Full guidelines are available on our Instructions for Authors page, http://www.life-sciencealliance.org/authors We encourage our authors to provide original source data, particularly uncropped/-processed electrophoretic blots and spreadsheets for the main figures of the manuscript. If you would like to add source data, we would welcome one PDF/Excel-file per figure for this information. These files will be linked online as supplementary "Source Data" files. ***IMPORTANT: It is Life Science Alliance policy that if requested, original data images must be made available. Failure to provide original images upon request will result in unavoidable delays in publication. Please ensure that you have access to all original microscopy and blot data images before submitting your revision.*** The manuscript of Dr P. Lefrançois et al., entitled: "The transcriptional landscape analysis of basal cell carcinomas reveals novel signaling pathways and actionable targets" presents analysis of publically available RNAseq data from 75 BCC and 34 samples of normal skin. Analyses of Lefrancois et al., tackled different aspects of BCC heterogeneity and reveal differentially expressed genes and pathways in BCC vs. skin and in BCC associated with aggressiveness.
Comments: One of the limitations of this analysis is associated with the very small number of samples representing rare subtypes. This has to be discussed in the text.
How did the authors assess the quality of the sequenced RNA? What was the Transcript Integrity Number (TIN) for each RNA sample, and was it taken into account? Samples from Atwood and Sharpe may be overlapping. Was uniqueness of the sample origin verified? In general, most aggressive tumors are frequently bigger and with better tumor purity (TP) than benign lesions, this may explain "enrichment" of tumor markers associated with aggressiveness. A graph representing distribution of tumor purity (TP) would be useful. It is also possible to compare TP estimates based on RNA with TP from mutational data. TP is an important confounding factor which can impact sample clusterization. PCA may help to reveal and account for the TP component. In figure 4c, high-risk BCC do not seem to form a separate cluster, low-risk BCC also represent one pure group, and other -includes both low and high risk BCC. There is no clear message about the differences at RNA level between these subgroups. How classification accuracy was calculated? Same comment is relevant for figure 5. One of the interesting findings is the upregulation of WNT pathway in vismo-resistant BCC. Would this result remain if vismo-resistant BCC contrasted with naïve to treatment BCC, majority of which are expected to respond to vismodegib? This would help to improve statistics. Labeling of the figures does not correspond to references in the text.

Reviewer #2 (Comments to the Authors (Required)):
This is a very interesting manuscript form an established group of investigators. The authors identified IL-17 pathways previously not shown in various clinical types of BCC. While there is a certain novelty of their research, there are several minor points that need to be addressed. p. 5, the second paragraph, about lipid-related metabolic pathways and "this can be reflected in the tendency for BCC to ulcerate and/or easily bleed." This statement is purely hypothetical and should be substantiated by the appropriate references. I would recommend moving it to a discussion from the results.
All violin plots are lacking statistical analysis of the differences. Please provide the appropriate statistics on the charts. p.6 "Validation in McGill BCC samples cohort." The authors identified 19 differentially-expressed genes that they try to validate. Traditionally, the validation involves ROC curves and AUC comparison between the previous set and new set of samples. It is tough to judge if those results are valid by a cluster analysis. p.7 "High risk BCC with histopathological subtype requiring Mohs..." "BCC tumors with superficial and nodular histopathological patterns are usually considered lowrisk." This statement is incorrect. The nodular BCC are never considered to be a low-risk, and for that reason, none of the topical treatments are FDA approved for the treatment of nodular BCC. There are very strict criteria for Mohs selection of NMSC, which include not only histological type (by the way, basosquamous BCC is not one of them, but metatypical and keratotic is included), but also area of the body, the patient characteristics, and the positive margin on the recent excision/biopsy (Connoly et all AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria... JAAD, 2012). Thus, it is not entirely clear to me what criteria were used for high-risk BCC samples. Samples that were obtained from Mohs? Some histological features of BCC? In any case, the samples may be too heterogenous, and I am questioning the utility of such separation. p.8 "Advances vs. non-advanced BCC" Please clarify the clinical and histological characteristics of your advanced BCC. What was the reason not to perform the surgery or radiation? Size of the tumors, age of the patients, location? All those features may contribute to the selection bias of your BCC in the advanced group.
p.9 how vismodegib-sensitive and resistant tumors were established? Ex vivo? Clinical response? please clarify p. 10 "for advanced BCC, anti-IL17 therapy may have activity..." the reference that was provided at the end of that sentence does not support this statement. Please, supply the correct reference.
p.11 discussion about natalizumab. I am not sure why there is a need to discuss JC virus reactivation in the manuscript about BCC. Figure 7A. The nodular BCC was shown on a histological picture, and the title underneath stated "conservative management (superficial BCC)". First of all, it is not a superficial BCC; second of all, the conservative management (whatever that means) is not appropriate for nodular BCC. The same goes for the next picture. The same BCC, the title underneath stated "simple BCC (superficial BCC)": again not superficial BCC on the picture but nodular and not sure what "simple" means. Please make some corrections.
Reviewer #3 (Comments to the Authors (Required)): Lefrancois et al. perform an in-depth analysis of gene expression of basal cell carcinomas (BCC) to reveal novel pathways that may serve as targets for future treatment. The investigators studied a number of different BCC variant transcriptomes that include low-risk, high-risk, advanced, vismodegib sensitive and vismodegib resistant tumors. The team analyzed differences in gene expression utilizing RNAseq from publically available dataset. The combination of datasets yield robust findings for the different variants of BCC that highlight pathways important in the pathogenesis of BCC. The authors analyzed gene sets for relevant pathways using KEGG, PANTHER, Reactome and BioCarta. The genes identified revealed pathways that included IL-17, TLR, Akt/PI3K, cadherins, integrins, and the Wnt/beta-catenin pathways. A strength of the study is that the investigators were able to validated a subset of genes by qRT-PCR from primary BCC samples. The analysis is novel and innovative, and new targets for treatment of BCC may potentially be developed.

Reviewers' comments:
Reviewer #1: We are thankful for the astute comments by Reviewer #1 that led us to resubmit an improved manuscript.
The transcriptional landscape analysis of basal cell carcinomas reveals novel signaling pathways and actionable targets" presents analysis of publically available RNAseq data from 75 BCC and 34 samples of normal skin. Analyses of Lefrancois et al., tackled different aspects of BCC heterogeneity and reveal differentially expressed genes and pathways in BCC vs. skin and in BCC associated with aggressiveness.

R1-C1
One of the limitations of this analysis is associated with the very small number of samples representing rare subtypes. This has to be discussed in the text.
Our response: We agree that this is a major limitation for analyses representing rare subtypes. We have added an additional paragraph describing this limitation.
"A major limitation of this study is the lack of power due to small sample size. This affects more strikingly uncommon groups of tumors, such as vismodegib-treated, vismodegib-sensitive tumors (n = 5) and BCC with a high-risk histopathological type (n = 6). Other groups had at least ten samples. In the hope of increasing robustness, for the comparison on vismodegib resistance, we also validated the major findings using the larger group of vismodegib-naïve BCCs as a control group, instead of vismodegib-treated, vismodegib-sensitive BCCs. Larger cohorts of patient samples might shed light on additional biologically meaningful pathways that could be missed in the current comparisons." Lines 276-283

R1-C2
How did the authors assess the quality of the sequenced RNA? What was the Transcript Integrity Number (TIN) for each RNA sample, and was it taken into account?
Our response: We have used standard RNA-Seq processing, including removal of low quality reads using FastX Toolkit as stated in the methods. The most commonly used method for RNA quality evaluation of RNA-Seq experiments is the experimental determination of RNA integrity numbers (RIN), but we do not have access to these data due to the design of our study. We had not initially computed Transcript Integrity Number (TIN) [Wang et al., 2016], since these were not reported in the original articles.
1st Authors' Response to Reviewers February 27, 2021 We have computed TINs for all samples. The median median TIN (medTIN) was 77.12 and the mean medTIN was 68.62. We have plotted medTINs on a per-sample basis, according to the 4 comparisons performed in the text, and assessed whether distributions differed between compared conditions using Mann-Whitney U tests (e.g. BCC vs. normal, advanced vs. non-advanced, …). No statistically significant differences were found. We have added a paragraph in the methods and a Supplementary Figure 6.
We did not account for TINs within our analyses, for 2 reasons. 1-We already accounted for batch effect and tumor purity in our analyses (see below). 2 -There are no consensus methods to account for TINs with wide acceptance/use. As such, it is not used by large consortia such as ENCODE. Some algorithms have started to emerge to correct for TINs (DegNorm, flexiMAP and others), but have not been widely adopted yet. We hope that there is a more streamlined approach in the future to account for TINs.

R1-C3
Samples from Atwood and Sharpe may be overlapping. Was uniqueness of the sample origin verified?
Our response: We were aware of this possibility and thus have verified the uniqueness of samples prior to analyses. We considered only non-overlapping samples in all our analyses.

R1-C4
In We appreciate the value in using mutational data to compare tumor purity.
Unfortunately, for most studies, mutational data were not readily available.
We applied the suggestion to account for tumor purity for clustering analyses, in addition to a batch effect. All heatmaps were updated to reflect these changes.
To better visualize effects from tumor purity, as suggested by Reviewer #1, we have performed Principal Component Analysis (PCA) on gene expression and tumor purity. We have generated PCA plots for all four comparisons (Supplementary Figure 8). Points on the PCA plots were sized according to tumor purity (higher tumor purity = larger size). We did not observe significant clustering of samples based primarily on tumor purity.   To better visualize effects from tumor purity, we have performed Principal Component Analysis (PCA) on gene expression and tumor purity. We have generated PCA plots for all four comparisons according to the first two principal components, with points sized according to tumor purity (higher tumor purity = larger size) (Supplementary Figure 8). We did not observe significant clustering of samples based primarily on tumor purity." Lines 371-387

R1-C5
In figure 4c, high-risk BCC do not seem to form a separate cluster, low-risk BCC also represent one pure group, and other -includes both low and high risk BCC. There is no clear message about the differences at RNA level between these subgroups. How classification accuracy was calculated? Same comment is relevant for figure 5.
Our response: For classification accuracy, we have used V-measure [Rosenberg et al., 2007], which encompasses homogeneity of clusters and completeness of clusters. We have added this metric to all heatmap analyses.
We agree with Reviewer #1 that there is a mixed cluster with high-risk and low-risk tumors and not a true pure cluster. This is also the case, as pointed out, for advanced BCC and non-advanced BCC. We have expanded the discussion of clustering results to mention similarities in RNA expression of few genes that account at least in part for these mixed clusters.
"Overall group membership was 93.5%, with a V-Measure of 0.48 (homogeneity = 0.66, completeness = 0.38)." Lines 82-83 "Hierarchical clustering on high-risk vs. low-risk BCC samples based on expression levels of all 67samples yielded 3 major patterns: one large cluster consisting only of low-risk BCCs, few early branching outgroups with high-risk BCCs, and a mixed cluster of high-risk and low-risk tumors ( Figure 4A). The mixed cluster is comprised of samples with overall moderate upregulation of SPHK1, MTHFD1, BMS1P20, PRMT6, ANGEL1, OLFML2B, and C1QTNF6, without any being highly upregulated. Clustering metrics were lower (V-measure = 0.28 with homogeneity = 0.52 and completeness = 0.19), while accurate group membership was 93.2%." Lines 157-163 "Hierarchical cluster analysis was performed using the expression levels for the top 50 upregulated genes. Three major groups were observed: one large cluster consisting purely of non-advanced BCCs, one early outgroup joined with a small cluster of three advanced BCCs, and a mixed cluster without strong sub-cluster associations between advanced and non-advanced tumors ( Figure 5A). The mixed cluster is comprised of samples with moderate to moderate-high upregulation of COL1A1, COL1A2, COL3A1, FN1, and LUM1, without a strong SFRP2 expression. Clustering metrics were lower (V-measure = 0.35 with homogeneity = 0.50 and completeness = 0.27), while accurate group membership was 87.3%." Lines 183-190 "Despite a limited number of differentially expressed genes, a group membership score of 100% was achieved when considering one large cluster only consisting of vismodegib-resistant tumors, and one looser association of an early outgroup of vismodegib-sensitive BCCs joined to a small cluster of vismodegib-sensitive BCCs. Clustering metrics were excellent, with a V-Measure of 0.84 (homogeneity = 1 and completeness = 0.72)." Lines 220-225 "V-measure, a clustering validity metric combining homogeneity of clusters and completeness of clusters, were determined for all heatmaps [Rosenberg et al., 2007]." Lines 408-410

One of the interesting findings is the upregulation of WNT pathway in vismoresistant BCC. Would this result remain if vismo-resistant BCC contrasted with naïve to treatment BCC, majority of which are expected to respond to vismodegib? This would help to improve statistics.
Our response: We are grateful for this idea that adds robustness to our manuscript. By contrasting vismodegib-naïve untreated BCC (most of them being low-risk BCC) and vismodegib-treated, vismodegib-resistant BCC, we confirmed that the Wnt/βcatenin pathway is also enriched among upregulated genes in vismodegib-resistant tumors (p = 0.016; Panther). The two novel Wnt/B-catenin genes upregulated when comparing resistant to sensitive tumors, DACT1 (Fold change=1.91; Q-value = 0.007) and FSTL1 (Fold change=1.92; Q-value = 0.0002), were also upregulated in vismodegib-resistant BCC when using naïve tumors as a control set. We have added a paragraph in the results section and discussed the findings in the discussion section as well. We have added a panel to Figure 6 to present violin plots for both genes (DACT1 and FSTL1) when comparing naïve to resistant tumors.

R1-C7
Labeling of the figures does not correspond to references in the text.
Our response: We are truly sorry for our mistake and apologize for any inconvenience to the editors and reviewers. We have re-ordered all figure panels according to the references in the text.

Reviewer #2:
We thank Reviewer #2 for bringing to our attention many essential points that needed clarifications and modifications.

This is a very interesting manuscript form an established group of investigators.
The authors identified IL-17 pathways previously not shown in various clinical types of BCC. While there is a certain novelty of their research, there are several minor points that need to be addressed.

R2-C1
p. 5, the second paragraph, about lipid-related metabolic pathways and "this can be reflected in the tendency for BCC to ulcerate and/or easily bleed." This statement is purely hypothetical and should be substantiated by the appropriate references. I would recommend moving it to a discussion from the results.
Our response: We agree with Reviewer #2 that the statement in the current form was too speculative. We have rewritten these sentences with better substantiation with current literature. The tendency for BCC to ulcerate and bleed might be better explained by increased microvessel density with increased angiogenesis due to inflammatory cytokines and VEGF [Aoki et al., 2003]. However, alterations in lipid metabolism are more frequently associated with non-healing wound.
-Moved section from Results to Discussion -It now reads: "Among BCC tumors, lipid-related metabolic pathways that are essential to generate, maintain and regenerate a fully functional keratinized epidermis appear to be downregulated. For example, impaired lipid metabolism and downregulation of PPAR pathways are associated with impaired wound healing [Michalik et al., 2006]. These alterations might contribute to the non-healing phenotype of certain BCCs." Lines 271-275

R2-C2
All violin plots are lacking statistical analysis of the differences. Please provide the appropriate statistics on the charts.
Our response: We have added FDR-adjusted p-values/Q-values for all Violin plots directly on the charts.

p.6 "Validation in McGill BCC samples cohort." The authors identified 19 differentially-expressed genes that they try to validate. Traditionally, the validation involves ROC curves and AUC comparison between the previous set and new set of samples. It is tough to judge if those results are valid by a cluster analysis.
Our response: We believe this point needs clarification. The heatmap presented as a validation in our cohort of BCC samples is a simplified visual representation of all RT-qPCR experiments rather than a clustering analysis, using RNA-Seq data as the discovery stage. Colors refer to fold-enrichment for that particular gene-sample combination, after normalization. There are obvious technical differences between RNA-Seq and RT-qPCR that make it challenging to directly compare results.
Using a simple approach where >50% of upregulated genes should have >2-fold enrichment to be classified as defining the BCC phenotype/disease, this method yielded an AUC of 0.933, sensitivity of 0.87 and specificity of 1.00. This is now added to a Supplementary Figure 3 -

panel A.
We have also added additional graphical representation of individual biomarkers as ROC curves, considering our cohort as a new set of samples. This is now Supplementary Figure 3 - We also included a mention that a larger validation cohort is needed to further characterize these biomarkers.
- Supplementary Figure 3 "Overall, 13 out of 15 BCC samples show normalized enrichment above 2-fold for at least 50% of qRT-PCR-tested genes (heatmap summarizing results in Figure 3); this simple approach yielded a sensitivity of 0.87 and a specificity of 1.0 for diagnosing a BCC (Supplementary Figure   3A; AUC = 0.93). Receiver-Operator Curves for individual genes as potential biomarkers are presented in Supplementary Figure 3B." Lines 119-124 "A larger validation cohort is likely needed to further characterize these biomarkers." Lines 343-344 "Receiver-Operator Curves (ROC) curves were generated using R package pROC [Robin et al., 2011]." Lines 424-425

R2-C4
p.7 "High risk BCC with histopathological subtype requiring Mohs..." "BCC tumors with superficial and nodular histopathological patterns are usually considered low-risk." This statement is incorrect. The nodular BCC are never considered to be a low-risk, and for that reason, none of the topical treatments are FDA approved for the treatment of nodular BCC. There are very strict criteria for Mohs selection of NMSC, which include not only histological type (by the way, basosquamous BCC is not one of them, but metatypical and keratotic is included), but also area of the body, the patient characteristics, and the positive margin on the recent excision/biopsy (Connoly et all AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria... JAAD, 2012). Thus, it is not entirely clear to me what criteria were used for high-risk BCC samples. Samples that were obtained from Mohs? Some histological features of BCC? In any case, the samples may be too heterogenous, and I am questioning the utility of such separation.
Our response: We thank Reviewer #2 for bringing to our attention these valid concerns. We have put greater emphasis on the wide clinical and histopathological presentations of BCC, on how features related to tumor characteristics, body area, and patient characteristics may portend a low risk or a high risk for local recurrence and aggressive tissue destruction, and on the importance of using the 2012 appropriate use criteria for Mohs. We have removed basosquamous and added keratotic from descriptions -none were represented among the sample set.
We have specified that, in this manuscript, we considered only the Mohs appropriate use criterion of "tumor characteristics -histopathological subtypes with aggressive features" [Ad Hoc Task force, Connolly et al., 2012] to define high-risk vs. low-risk BCC. This statement was included several times in the manuscript as a reminder.
We see value of this separation as a proxy for intrinsic biological behavior. Genomic studies on rare histological BCC subtypes are uncommon and we believe they can bring meaningful data to understand their underlying pathogenesis and guide their management in the future (alternative treatment options, combination with Mohs and medical therapies, etc.).
We have erased "Conservative" vs. "Mohs" on figures to remove any confusion.
"BCC tumors can present with a wide range of clinical and histopathological presentations, which may render them high risk for local recurrence and aggressive tissue destruction. Features such as body locations, patient characteristics (immunosuppression, genetic syndromes, radiotherapy, etc.) and tumor characteristics (histopathological subtypes with aggressive features, margin positivity, recurrent tumor, etc.) all have the potential to influence BCC risk level. The 2012 American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and the American Society for Mohs Surgery appropriate use criteria for Mohs micrographic surgery help identify clinical scenarios most likely to benefit from Mohs surgery, in the context of limited accessibility [Ad Hoc Task force, Connolly et al., 2012]. BCC with high risk features are usually managed by Mohs micrographic surgery, which was shown to significantly reduce the recurrence rate over more traditional medical and/or surgical approaches [16].
For subsequent analyses, in this manuscript, we consider only the Mohs appropriate use criterion of "tumor characteristicshistopathological subtypes with aggressive features" [Ad Hoc Task force, Connolly et al., 2012]. Thus, BCC tumors with superficial and nodular (especially on a low risk body area) patterns would be considered low-risk, when compared to high-risk BCC tumors with histological subtypes indicating a more aggressive biological behavior, such as morpheaform, infiltrating, metatypical, keratotic and micronodular.

R2-C5
p.8 "Advances vs. non-advanced BCC" Please clarify the clinical and histological characteristics of your advanced BCC. What was the reason not to perform the surgery or radiation? Size of the tumors, age of the patients, location? All those features may contribute to the selection bias of your BCC in the advanced group Our response: Detailed clinical information were limited in the 3 main studies from which RNA-Seq data originated. All advanced tumors satisfied the criteria that were used for FDA approval of vismodegib and sonidegib as we reported below: "Advanced BCCs include locally advanced and metastatic tumors……. Although locally advanced BCCs do not have formal definite criteria, they represent tumors for which surgery and/or radiotherapy are usually not curative or not appropriate (e.g. resulting in mutilation) [9]." Lines 174-178 We added a statement that all advanced BCCs satisfied these criteria: "All advanced tumors satisfied the inclusion criteria given in the prior paragraph." Lines 179-

180
The precise information why radiation or surgery was not performed on a per patient basis remains unknown. We have added a cautionary statement in the discussion about this limitation: "Another limitation relates to missing key elements from the clinical information. For example, for advanced BCCs, it was not possible to know whether a selection bias might have been a reason why surgery and/or radiation were not performed, or whether patient and tumor factors such as the underlying location, immunosuppression status, and histopathological subtypes played a role in clinical decision making." Lines 284-288

R2-C6
p.9 how vismodegib-sensitive and resistant tumors were established? Ex vivo? Clinical response? please clarify Our response: Clinical response was used to establish vismodegib resistance and sensitivity, as per RECIST guidelines (progressive disease or stable disease for resistance, any responder for sensitivity). We added this key information in the results section: "Clinical response to vismodegib as defined by the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines was the endpoint for vismodegib sensitivity (partial response or complete response) vs. resistance (progressive disease or stable disease) [Atwood et al, 2015][Sharpe et al., 2015." Lines 214-216

R2-C7
p. 10 "for advanced BCC, anti-IL17 therapy may have activity..." the reference that was provided at the end of that sentence does not support this statement. Please, supply the correct reference.
Our response: We apologize that the end of the statement and the reference did not match the initial part. We have corrected the statement and added the relevant references. We also expanded the section on IL-17 as a therapeutic target for advanced BCC with relevant studies in animals and in cell lines. We added that this remains to be proven in clinical trials.
"IL-17 and IL-22 cytokines have been directly implicated in BCC and SCC progression and tumor growth in mouse xenografts, as well as progression, cell migration and local invasion in BCC cancer cell lines [Nardinocchi et al., 2015]. Another report confirmed the upregulation of both cytokines in peritumoral skin of BCC, correlating with the severity of the inflammatory infiltrate [Pellegrini et al., 2017]. Both studies have suggested IL-17 as a potential therapeutic target. From our analysis, for advanced BCC, there might be a potential benefit using anti-IL-17 therapies for treating these tumors, with a favorable safety profile. This effect remains to be proven in clinical trials." Lines 257-264

R2-C8
p.11 discussion about natalizumab. I am not sure why there is a need to discuss JC virus reactivation in the manuscript about BCC Our response: We have removed the section on Natalizumab given its lesser relevance to BCC as mentioned by Reviewer #2. Figure 7A. The nodular BCC was shown on a histological picture, and the title underneath stated "conservative management (superficial BCC)". First of all, it is not a superficial BCC; second of all, the conservative management (whatever that means) is not appropriate for nodular BCC. The same goes for the next picture. The same BCC, the title underneath stated "simple BCC (superficial BCC)": again not superficial BCC on the picture but nodular and not sure what "simple" means. Please make some corrections.

R2-C9
Our response: We appreciate the astute observations brought by Reviewer #2. As discussed above, we have removed Conservative management and Mohs, and replaced them by low-risk BCC and high-risk BCC. We have replaced the picture of a nodular BCC by the picture of a superficial BCC. We agree that "Simple" BCC does not convey much meaning and is incorrect. We have changed "Simple" to nonadvanced BCC to encompass BCC that do not satisfy the definition of advanced BCC. We have updated the Figure accordingly.
-Updated Figure 7A Reviewer #3: We are glad Reviewer #3 enjoyed our manuscript and found our approach useful to identify actionable targets for BCC.