Target Discovery in Head-and-Neck Squamous Cell Carcinoma: Genome-Wide CRISPR Screens Illuminate Therapeutic Resistance and Actionable Dependencies

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Overall comments:   This review of studies that leverage state-of-the-art CRISPR screens to identify targets and dependencies in HNSCC is in principal a useful resource and good starting place for those who are new to this technology and its history, particularly in the niche of HNSCC. However, the review needs to be updated and move beyond superficial summaries as pointed out below.  If the authors can correct these comments, the review will be greatly improved.

Abstract and Text

1. The statement that DepMAp included 34 OCSCC HNSCC cell lines is likely incorrect. In more recent data updates, I counted 58 oral cavity cell lines, 14 laryngeal/hypopharyngeal cell lines, 7 undesignated subtypes, plus one oropharyngeal cell lines for a total of 81 HNSCC cell lines with DepMap data. Making HNSCC the most represented cancer in the DepMap Project.   This should also be updated in the text in many other places where the reference is to what DepMAP did in 2019, as there are much newer database releases with more HNSCC cell lines screened. Please double check.

Text

2. The authors bring up the study by Chai et al that supports YAP1 as an essential gene in a smaller 14 panel HNSCC cell line cohort screened, but this interpretation seems unlikely given the redundancy between YAP1 and TAZ, and it is more likely that some of the cell lines tested showed dependency on YAP1, but levels are not stated in this review nor what kind of metric was used. Generally, to conclude a gene is “essential” has certain implication regarding both the % of cell lines tested that were effected and the metric of dropout or degree of dependency. Without discussing these nuances, the text may be misleading.
3. If the authors intend to review individual studies then it would behoove them to compile a master table of how all the dependent genes among these different studies actually overlaps in venn diagram or box plots (in case there are more than 4 studies to cross compare). I would suggest breaking this down into screens that looked at baseline dependency and then screens that looked at stress induced dependency and dig through the data to see if there is more overlap than perhaps the individual genes highlighted in each study. Otherwise, the review remains a bit superficial.
4. To move beyond the flavor of just providing a list of references, the reviewers should provide a bit more detail of the robustness of each study mentioned. For example, the study by Li et al identified STK19 KO as synthetic lethal with cisplatin treatment but did the authors of that study restrict their validation to just one or two cell lines screened or was it tested more robustly in a larger panel of HNSCC cell lines. Otherwise the point of this review seems to only make us aware of what is possible, rather than offering insight into what has been discovered.
5. In the text referring to the screen by Goto and Colleagues, the text says the screens were designed to identify genes whose loss enhances mTOR inhibitor killing. Please check if the screen phenotype was cell death verses drop out, which is a combined phenotype that could be driven largely by proliferation. Clarify what kind of screen this was and then perhaps if individual validation studies showed an increase in cell death markers beyond an increased anti-proliferative effect.

 

Comments on the Quality of English Language

Needs just a few grammatical corrections here and there, but I could understand the ideas well enough.

Author Response

Overall comments: This review of studies that leverage state-of-the-art CRISPR screens to identify targets and dependencies in HNSCC is in principal a useful resource and good starting place for those who are new to this technology and its history, particularly in the niche of HNSCC. However, the review needs to be updated and move beyond superficial summaries as pointed out below. If the authors can correct these comments, the review will be greatly improved.

Abstract and Text

1. The statement that DepMAp included 34 OCSCC HNSCC cell lines is likely incorrect. In more recent data updates, I counted 58 oral cavity cell lines, 14 laryngeal/hypopharyngeal cell lines, 7 undesignated subtypes, plus one oropharyngeal cell lines for a total of 81 HNSCC cell lines with DepMap data. Making HNSCC the most represented cancer in the DepMap Project. This should also be updated in the text in many other places where the reference is to what DepMAP did in 2019, as there are much newer database releases with more HNSCC cell lines screened. Please double check.

Response: We thank the reviewer for catching this oversight and apologize for the mistake. We re-checked the current DepMap release (accessed on [12/2/2025]) and confirm that HNSCC coverage is 89 cell lines (including oral cavity, laryngeal/hypopharyngeal, oropharyngeal, and undesignated subtypes), rather than the 34 reported from older 2019 snapshots. We have replaced “34” with “89” throughout the manuscript and updated the language to reflect the current release, noting that counts may change with future updates. The amendments can be found on page 1 (line 21), page 3 (line 92), and page 11 (line 143 and 163).

2. The authors bring up the study by Chai et al that supports YAP1 as an essential gene in a smaller 14 panel HNSCC cell line cohort screened, but this interpretation seems unlikely given the redundancy between YAP1 and TAZ, and it is more likely that some of the cell lines tested showed dependency on YAP1, but levels are not stated in this review nor what kind of metric was used. Generally, to conclude a gene is “essential” has certain implication regarding both the % of cell lines tested that were effected and the metric of dropout or degree of dependency. Without discussing these nuances, the text may be misleading.

Response: We thank the reviewer for highlighting these nuances to avoid any misleading issues in the write-up. We have made the amendment accordingly on page 11 from lines 149 to 161.

3. If the authors intend to review individual studies then it would behoove them to compile a master table of how all the dependent genes among these different studies actually overlaps in venn diagram or box plots (in case there are more than 4 studies to cross compare). I would suggest breaking this down into screens that looked at baseline dependency and then screens that looked at stress induced dependency and dig through the data to see if there is more overlap than perhaps the individual genes highlighted in each study. Otherwise, the review remains a bit superficial.

Response: We thank the reviewer for this constructive suggestion. We agree that integrating the dependency genes from all published studies that involve CRISPR screens into a unified analytical framework such as a Venn diagram or clustering of baseline versus stress-induced dependencies would further enhance understanding of shared vulnerabilities in HNSCC. However, the current work is intended as a narrative review rather than a re-analysis of primary screening data. The summarized studies differ evidently in design screening library composition (GeCKO v1/v2, Brunello, kinome-focused, or custom), MOI and perturbation context (baseline proliferation versus cisplatin, radiation, or oncolytic HSV-1 exposure). Therefore, direct numerical overlap cannot be assessed without regularizing raw sgRNA-count data across platforms, which is beyond the scope of this review.

To address this feedback, we classified each study according to its screening context, whether designed to identify baseline essential genes or treatment-perturbed dependencies such as those induced by cisplatin, radiation, or targeted therapy or oncolytic virus in the table 1. The table legend now reads as follows: Summary of sgRNA libraries and experimental approaches in CRISPR Cas9-screening studies of HNSCC. The table presents the screening context (either baseline or treatment perturbation), findings, and limitations, highlighting translational opportunities and current gaps across various studies that involve chemotherapy, radiation, targeted, metabolic, and oncolytic contexts.

We also have added a write up in the manuscript to acknowledge the divergency of screening approaches and to highlight the significance of future cross-study meta-analyses to determine essential genes and context-specific vulnerabilities under the section on future studies on page 14 (line 309-320).

With this amendment, we hope the clarity and depth of the manuscript have been improved while its narrative focus has been maintained.

4.To move beyond the flavor of just providing a list of references, the reviewers should provide a bit more detail of the robustness of each study mentioned. For example, the study by Li et al identified STK19 KO as synthetic lethal with cisplatin treatment but did the authors of that study restrict their validation to just one or two cell lines screened or was it tested more robustly in a larger panel of HNSCC cell lines. Otherwise the point of this review seems to only make us aware of what is possible, rather than offering insight into what has been discovered.

Response: We appreciate this insightful suggestion. The section discussing Li et al. (2025) has been revised to include additional details on the robustness and validation scope of the study. The amendments can be found on page 12 (line 198-211). Specifically, we now clarify that the CRISPR dropout screen was performed across two independent tongue SCC cell lines (TSCCA and CAL27), followed by both genetic (STK19 knockout) and pharmacologic (T-12-037-01 inhibitor) validation. Furthermore, we highlight that the in vivo combination of cisplatin and STK19 inhibition significantly suppressed TSCCA-derived tumor growth, confirming the findings beyond a single model.

5.In the text referring to the screen by Goto and Colleagues, the text says the screens were designed to identify genes whose loss enhances mTOR inhibitor killing. Please check if the screen phenotype was cell death verses drop out, which is a combined phenotype that could be driven largely by proliferation. Clarify what kind of screen this was and then perhaps if individual validation studies showed an increase in cell death markers beyond an increased anti-proliferative effect.

Response: We thank the reviewer for this comment. The section referring to Goto et al. (2024) has been revised to specify that the study utilized a kinome-wide CRISPR-Cas9 dropout (loss-of-function) assay, in which sgRNA depletion over time reflected reduced proliferative fitness rather than direct cytotoxicity. We have also noted that the study did not investigate apoptosis or other cell-death markers beyond anti-proliferative effects. The revised text now accurately represents the experimental design and findings as reported by the authors. The amendments can be found on page 13 (line 251-269).

6. Needs just a few grammatical corrections here and there, but I could understand the ideas well enough.

Response: We thank the reviewer for the positive feedback and kind remark. The manuscript has been carefully revised to correct minor grammatical issues and improve overall readability. Additionally, it has been subjected to professional English editing by a native English speaker through the MDPI Author Services to ensure clarity and linguistic accuracy.

 

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

The author conducted a review focusing on the use of genome-wide CRISPR screening to identify essential genes, mechanisms of therapy resistance, and synthetic-lethal vulnerabilities in head and neck squamous cell carcinoma, with the aim of advancing precision therapeutics beyond current limited treatments such as cetuximab and PD-1 inhibitors. The integration of all these themes appears to be original in the current literature. The manuscript is well written, and the author employed a reproducible and systematic search strategy. For these reasons, I believe the manuscript is suitable for publication in Biomedicines. As a minor suggestion, I recommend that the author include graphs and diagrams to better illustrate the relationships among the investigated variables and make the article more visually engaging for readers.

Author Response

The author conducted a review focusing on the use of genome-wide CRISPR screening to identify essential genes, mechanisms of therapy resistance, and synthetic-lethal vulnerabilities in head and neck squamous cell carcinoma, with the aim of advancing precision therapeutics beyond current limited treatments such as cetuximab and PD-1 inhibitors. The integration of all these themes appears to be original in the current literature. The manuscript is well written, and the author employed a reproducible and systematic search strategy. For these reasons, I believe the manuscript is suitable for publication in Biomedicines. As a minor suggestion, I recommend that the author include graphs and diagrams to better illustrate the relationships among the investigated variables and make the article more visually engaging for readers.

Response: We thank the reviewer for this constructive suggestion. In response, we have added a new graphical abstract (now included as Figure 1) on page 10 that visually summarizes the overall concept of genome-wide CRISPR/Cas9 dropout and treatment-perturbed screening strategies in head and neck squamous cell carcinoma (HNSCC).

Reviewer 3 Report

Comments and Suggestions for Authors

The author presents a comprehensive review on the use of genome-wide CRISPR/Cas9 screens to identify therapeutic vulnerabilities and resistance mechanisms in HNSCC. I believe the manuscript addresses a highly relevant and timely topic, given the growing interest in leveraging functional genomic platforms to accelerate the discovery of actionable targets in aggressive tumors. Below, I offer several comments that may help strengthen the review.

In the section on research strategy, the author states that the search covered the period 2021–2025; however, key studies from 2019 are discussed. I recommend adjusting the timeframe to a more realistic range or clarifying explicitly why studies prior to the stated period were included.

Overall, the author summarizes the findings of each study appropriately; however, it would be helpful to add, at the end of each subsection, a short integrative paragraph that highlights recurrent pathways or shared biological themes.

The author notes in several sections that previous studies lack in vivo validation. I agree with this point, but I suggest avoiding repetition in every study summary. It may be more effective to consolidate this limitation into a single paragraph at the end of Sections 3.1 or 3.2, indicating that the absence of in vivo validation represents a systematic gap in the current literature.

In Section 3.2, the writing includes subjective expressions such as “Sadly, the author did not…”. I recommend replacing these with more neutral academic phrasing (“However, the study did not evaluate…”).

Table 1 contains a large amount of text under the “Findings” column. I recommend using bullet points, standardizing the type of information reported across studies, and, if possible, adding key limitations. In this regard, avoiding long narrative descriptions may also help the table read more clearly.

The “Future Directions” section has substantial potential. I suggest adding a clearer prioritization of the areas in which CRISPR-based approaches could genuinely transform HNSCC management.

Lastly, I recommend a brief grammatical review, as there are some vague or overly complex sentences throughout the text.

Author Response

1. In the section on research strategy, the author states that the search covered the period 2021–2025; however, key studies from 2019 are discussed. I recommend adjusting the timeframe to a more realistic range or clarifying explicitly why studies prior to the stated period were included.

Response: Thank you for noting this important point. The text in the Research Strategy section has been revised to clarify that earlier studies were included intentionally to provide foundational context for genome-wide CRISPR screening in HNSCC. The timeframe has been expanded to reflect this broader inclusion range. The amendments can be found on page 10 (line 119-126).

 

2. Overall, the author summarizes the findings of each study appropriately; however, it would be helpful to add, at the end of each subsection, a short integrative paragraph that highlights recurrent pathways or shared biological themes.

Response: Thank you for this valuable suggestion. We have added integrative summary paragraphs at the end of each subsection (Sections 3.1–3.3) to highlight recurrent biological pathways and shared mechanistic themes among the studies. The amendments can be found on page 12 (line 177-183), page 13 (line 232-246), page 14 (line 299-307).

 

 

3. The author notes in several sections that previous studies lack in vivo validation. I agree with this point, but I suggest avoiding repetition in every study summary. It may be more effective to consolidate this limitation into a single paragraph at the end of Sections 3.1 or 3.2, indicating that the absence of in vivo validation represents a systematic gap in the current literature.

Response: We appreciate this constructive observation. To reduce redundancy, we have removed repeated mentions of “lack of in vivo validation” across individual study descriptions and consolidated this limitation into a single integrative paragraph at the end of Section 3.2 on page 13 (line 232-240). This new paragraph summarizes that the absence of in vivo validation represents a systematic limitation in current CRISPR-based HNSCC studies, hindering the translation of candidate targets into preclinical and clinical applications.

 

 

 

 

4. In Section 3.2, the writing includes subjective expressions such as “Sadly, the author did not…”. I recommend replacing these with more neutral academic phrasing (“However, the study did not evaluate…”).

Response: Thank you for noting this. We have revised subjective or informal expressions in Section 3.2 to adopt neutral, academic language. For instance, phrases such as “Sadly, the author did not…” have been replaced with “However, the authors did not evaluate…” on page 13, line 228.

 

5. Table 1 contains a large amount of text under the “Findings” column. I recommend using bullet points, standardizing the type of information reported across studies, and, if possible, adding key limitations. In this regard, avoiding long narrative descriptions may also help the table read more clearly.

Response: We appreciate this suggestion to improve clarity and visual readability. Table 1 has been completely reformatted using concise bullet points under the “Findings” column. The information has been standardized across studies to consistently include the following: Key dependency or resistance genes that identified vulnerabilities; Findings in bullet points; Notable limitations (absence of in vivo validation and mechanistic link to down-stream effectors not tested). This revised format enhances readability and comparability between studies. The amendments can be found in table 1 on page 4.

 

 

6. The “Future Directions” section has substantial potential. I suggest adding a clearer prioritization of the areas in which CRISPR-based approaches could genuinely transform HNSCC management.

Response: Thank you for recognizing the strength of this section. We have reorganized the “Future Directions” section to clearly prioritize transformative areas for CRISPR-based approaches in HNSCC. Specifically, we now highlight the identification of chemoradiation sensitizer targets relevant to perioperative and neoadjuvant settings as the top priority, followed by functional genomic integration with single-cell and multi-omic datasets to uncover context-specific vulnerabilities. This is then followed by pooled in vivo and organoid-based CRISPR screening to capture immune and stromal interactions and, finally, the development of biomarker-driven CRISPR validation pipelines to accelerate translation toward precision therapy. This revised structure provides a more strategic roadmap for how CRISPR-based technologies could advance both target discovery and therapeutic development in HNSCC. The amendments can be found on page 15-16.

 

7. Lastly, I recommend a brief grammatical review, as there are some vague or overly complex sentences throughout the text.

Response: We thank the reviewer for the positive feedback and kind remark. The manuscript has been carefully revised to correct minor grammatical issues and improve overall readability. Additionally, it has been subjected to professional English editing by a native English speaker through the MDPI Author Services to ensure clarity and linguistic accuracy.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors made the changes suggested by the reviewers