Review Reports - Early-Phase Drug Development in Rare Cancers: A Decade of Progress and Future Directions

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a comprehensive narrative review summarizing the past‑decade advances of early‑phase drug development for rare cancers. The manuscript systematically describes innovative trial frameworks, multiple emerging therapeutic modalities, predictive biomarkers, global regulatory‑collaborative ecosystems, and future research roadmaps. Three informative figures and two summary tables well organize the large body of literature. The topic fits perfectly within the scope of Cancers. Overall, the review is well‑structured and informative, yet several conceptual, organizational and presentation issues need revision prior to acceptance.

Major:

Narrative‑review methodology description is too brief. The authors state a literature search using PubMed/MEDLINE and Google Scholar, without specifying search terms, date range, inclusion/exclusion criteria, or risk‑of‑bias assessment. Please expand this section to improve reproducibility of this narrative review.
Limited critical synthesis between tables and main text. Table 1 summarizes key tumor‑agnostic trials and Table 2 lists biomarker‑driven strategies. However, the main text rarely cross‑refers to these tables. Integrate in‑text citations pointing readers to Table 1 / Table 2 where relevant, and remove redundant duplicate descriptions between text and tables.
Gaps regarding real‑world‑evidence (RWE) limitations. The manuscript highlights benefit of real‑world data, synthetic control arms and external controls for ultra‑rare cancers, yet insufficiently discusses their major caveats: selection bias, missing confounding variables, heterogeneous data quality, and regulatory hesitations for pivotal approval. Add balanced discussion of these drawbacks.
Uneven depth across therapeutic classes. Cellular therapies section covers CAR‑T, TIL, TCR‑T, CAR‑NK and oncolytic viruses. However, there is little discussion about key barriers limiting solid‑tumor cellular‑therapy adoption specifically for rare‑cancer populations, such as limited access to manufacturing, high cost, and biospecimen scarcity. Please enrich this part.
Future directions section could strengthen actionable clinical‑research priorities. Current future directions cover diagnostics, resistance mechanisms, regulation and global collaboration. It would benefit from adding concrete, patient‑centric priorities: equity of genomic testing access, decentral/digital trial implementation, and patient‑advocacy engagement for ultra‑rare tumor subgroups.

Minor:

Typographical & formatting issues

Multiple long reference lists inside Table 1 show garbled line breaks. Clean table formatting for readability.
Several abbreviations first appear without full expansion (e.g., IRCI, ROAF). Add full name on first mention.

Figure‑related points

Figures 1‑3 are generated via BioRender. Confirm proper copyright attribution is complete for publication.
Figure 3 legend may briefly note that this collaborative ecosystem applies especially for ultra‑rare tumor indications.

Reference quality and consistency

Some in‑text citations point to review papers while primary pivotal trial reports are available. Where space permits, prioritize citing key primary clinical studies.
Check reference formatting consistency (journal abbreviations, doi placement).

Terminology clarification

Distinguish clearly between “tumor‑agnostic approval” and “tumor‑agnostic clinical activity”. Some agents show pan‑tumor efficacy but lack formal tissue‑agnostic regulatory approval (e.g., KRAS‑G12C inhibitors). Clarify this distinction in text and Table 1.

Author Response

Comment 1:

Response 1: We sincerely thank the reviewer for the time and care devoted to evaluating our manuscript, and for the encouraging comments on its scope, structure, and figures. We have addressed each of the points raised below.

Major:

Comment 2:

Narrative‑review methodology description is too brief. The authors state a literature search using PubMed/MEDLINE and Google Scholar, without specifying search terms, date range, inclusion/exclusion criteria, or risk‑of‑bias assessment. Please expand this section to improve reproducibility of this narrative review.

Response 2: We thank the reviewer for this constructive suggestion. As this is a narrative rather than a systematic review, we did not apply formal systematic-review procedures such as a risk-of-bias assessment. However, we have expanded the methodology section to describe our search strategy and the major databases searched in more detail, to improve transparency and reproducibility.

Comment 3:

Limited critical synthesis between tables and main text. Table 1 summarizes key tumor‑agnostic trials and Table 2 lists biomarker‑driven strategies. However, the main text rarely cross‑these tables. Integrate in‑text citations pointing readers to Table 1 / Table 2 where relevant, and remove redundant duplicate descriptions between text and tables.

Response 3: We thank the reviewer for this helpful comment. We have added in-text citations to Table 1 and Table 2 where relevant and removed descriptions that were duplicated between the text and the tables. We have also expanded the subsection on resistance mechanisms to provide a more critical synthesis of emerging therapies.

Comment 4:

Gaps regarding real‑world‑evidence (RWE) limitations. The manuscript highlights benefit of real‑world data, synthetic control arms and external controls for ultra‑rare cancers, yet insufficiently discusses their major caveats: selection bias, missing confounding variables, heterogeneous data quality, and regulatory hesitations for pivotal approval. Add balanced discussion of these drawbacks.

Response 4: We thank the reviewer for highlighting this important point. We have revised the discussion of external controls to address these points. We now distinguish contextual external comparisons from randomized evidence and clarify that their regulatory utility is case-specific.

Comment 5:

Uneven depth across therapeutic classes. Cellular therapies section covers CAR‑T, TIL, TCR‑T, CAR‑NK and oncolytic viruses. However, there is little discussion about key barriers limiting solid‑tumor cellular‑therapy adoption specifically for rare‑cancer populations, such as limited access to manufacturing, high cost, and biospecimen scarcity. Please enrich this part.

Response 5: We agree with the reviewer and thank them for this suggestion. We have expanded this section to discuss the biological barriers to cellular therapy in solid tumors (antigen heterogeneity, on-target toxicity, trafficking, stroma, and immunosuppression) together with the practical barriers that are particularly relevant to rare cancers (biopsy scarcity, manufacturing time and failures, cost, travel, and access). We now also recommend that studies report screening-to-infusion attrition in addition to outcomes among infused patients.

Comment 6:

Future directions section could strengthen actionable clinical‑research priorities. Current future directions cover diagnostics, resistance mechanisms, regulation and global collaboration. It would benefit from adding concrete, patient‑centric priorities: equity of genomic testing access, decentral/digital trial implementation, and patient‑advocacy engagement for ultra‑rare tumor subgroups.

Response 6: We thank the reviewer for these valuable suggestions. We have added concrete, patient-centered priorities to the Future Directions section. To address equity and the implementation of digital trials, we also moved the paragraphs on decentralized trials and advocacy groups from Section 6 to the Future Directions section.

Minor:

Comment 7:

Typographical & formatting issues

Multiple long reference lists inside Table 1 show garbled line breaks. Clean table formatting for readability.
Several abbreviations first appear without full expansion (e.g., IRCI, ROAF). Add full name on first mention.

Response 7: We thank the reviewer for these helpful suggestions. To improve formatting and readability, Table 1 has been split into three tables:

Table 1a. Cross-Tumor Biomarker-Driven Trials for Immunotherapy Relevant to Rare Cancers
Table 1b. Cross-Tumor Biomarker-Driven Trials for Targeted Therapies Relevant to Rare Cancers
Table 1c. Cross-Tumor Biomarker-Driven Trials for Fusion Inhibitors Relevant to Rare Cancers

All abbreviations, including those noted by the reviewer, have been checked and corrected.

Comment 8:

Figure‑related points

Figures 1‑3 are generated via BioRender. Confirm proper copyright attribution is complete for publication.
Figure 3 legend may briefly note that this collaborative ecosystem applies especially for ultra‑rare tumor indications.

Response 8: We thank the reviewer for raising these points. Each figure created with BioRender is covered by an appropriate publication license, and the copyright attribution appears beneath each figure description. We have also noted in the Figure 3 legend that this collaborative ecosystem applies especially to ultra-rare tumor indications.

Comment 9:

Reference quality and consistency

Some in‑text citations point to review papers while primary pivotal trial reports are available. Where space permits, prioritize citing key primary clinical studies.
Check reference formatting consistency (journal abbreviations, doi placement).

Response 9: We thank the reviewer for this suggestion. Primary pivotal trial reports and other primary clinical studies have been included where appropriate. The references now follow the recommended format with MEDLINE journal abbreviations, and DOIs that were missing from some references have been added.

Comment 10:

Terminology clarification

Distinguish clearly between “tumor‑agnostic approval” and “tumor‑agnostic clinical activity”. Some agents show pan‑tumor efficacy but lack formal tissue‑agnostic regulatory approval (e.g., KRAS‑G12C inhibitors). Clarify this distinction in text and Table 1.

Response 10: We thank the reviewer for this important clarification. Table 1 is now labeled to reflect cross-tumor, biomarker-driven trials relevant to rare cancers (with separate tables for immunotherapy, targeted therapies, and fusion inhibitors), and it states the FDA tumor-agnostic approval status of each agent or notes where no approval has been obtained. We have made corresponding changes in the text.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript effectively catalogs trial designs, therapies, and biomarkers but often comes across as a listing rather than a critical scholarly review. For example, Sections 4, 5, and 6 list numerous approvals and technologies sequentially without thoroughly examining the limitations of current evidence, including the statistical challenges of basket and umbrella trials, selection biases, response durability, failures of tumor-agnostic approaches, health-economic issues, reimbursement hurdles, and disparities in biomarker testing. A comprehensive review should address key questions such as why some tissue-agnostic therapies vary in effectiveness across different lineages, the limitations of ORR-based approvals, challenges in ultra-rare indications, and regulatory uncertainties linked to small single-arm datasets.

Figures 1 and 2 are visually engaging but very schematic. They oversimplify trial design principles, omit statistical considerations, and do not discuss operational complexities. Figure 1 particularly emphasizes AI and future directions without providing scientific support.

Several sections cover AI-driven multiomics, computational drug design, spatial transcriptomics, organoids, and digital trials. However, evidence supporting their use in rare cancers is still limited.

The review primarily highlights successful approvals and positive studies. A critical review should address unsuccessful basket trials, failures of biomarker-driven approaches, negative immunotherapy studies, resistance mechanisms that lead to clinical failure, and lessons learned from unsuccessful development programs.

The data and information provided, including those in the tables, must be thoroughly checked and validated. For example, the manuscript states that rare cancers account for 20-25% of diagnoses and 25-30% of cancer deaths. These figures vary throughout the manuscript and should be consistent with the cited literature.

The adaptive trial section discusses Bayesian and adaptive randomization principles but offers limited details on type I error control, multiplicity, cross-cohort information sharing, and regulatory issues.

Author Response

Comment 1:

Response 1: We thank the reviewer for this insightful critique. We have added a synthesis of the statistical and operational limitations of master protocols and of the constraints of ORR-based endpoints and small single-arm cohorts, drawing on evidence from SHIVA, the negative NCI-MATCH substudies, and the variable responses observed in SARC028. We also discuss BRAF V600E colorectal cancer as a mechanistic example of lineage-dependent resistance. These examples are interpreted within the design and sample-size limitations of each study.

Comment 2:

A comprehensive review should address key questions such as why some tissue-agnostic therapies vary in effectiveness across different lineages, the limitations of ORR-based approvals, challenges in ultra-rare indications, and regulatory uncertainties linked to small single-arm datasets.

Response 2: We thank the reviewer for these important questions. We have expanded the revised manuscript to address each of these points, as described in our Responses 1 and 5.

Comment 3:

Response 3: We thank the reviewer for this comment. We have revised the legends of Figures 1 and 2 to state that the figures are conceptual and to identify the statistical and operational elements omitted from the schematics. We have also revised Figure 1 to place less emphasis on A.I.

Comment 4:

Response 4: We agree with the reviewer. We have tempered the discussion of AI, spatial transcriptomics, and organoids to distinguish hypothesis-generating and preclinical work from prospectively validated clinical tools.

Comment 5:

Response 5: We thank the reviewer for this valuable point. We have added a new subsection (3.4. Statistical and Operational Challenges of Clinical Trial Design) as well as added additional details for resistance mechanisms to address a more critical review.

Comment 6:

Response 6: We thank the reviewer for pointing this out. The statistics on rare cancers throughout the manuscript have been cross-checked and updated so that they are consistent with the values reported by the International Agency for Research on Cancer (IARC) and the World Health Organization (WHO).

Comment 7:

Response 7: We thank the reviewer for this suggestion. In response, we have added a dedicated subsection (3.4. Statistical and Operational Challenges of Clinical Trial Design) to address this.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors address most of the concerns adequately.