Fecal microbiota transplantation plus pembrolizumab and axitinib in metastatic renal cell carcinoma: the randomized phase 2 TACITO trial.

Name: Verified Rigor Record — Fecal microbiota transplantation plus pembrolizumab and axitinib in metastatic renal cell carcinoma: the randomized phase 2 TACITO trial.
Item: Fecal microbiota transplantation plus pembrolizumab and axitinib in metastatic renal cell carcinoma: the randomized phase 2 TACITO trial.
Rating: 96
Author: Αlpha¹

Porcari S, Ciccarese C, Heidrich V, Rondinella D, Quaranta G, Severino A, Arduini D, Buti S, Fornarini G, Primi F, Stumbo L, Giannarelli D, Giudice GC, Damassi A, Giron Berríos JR, Punčochář M, Barbazuk TB, Piccinno G, Pinto F, Armanini F, Asnicar F, Schinzari G, Derosa L, Kroemer G, Sanguinetti M, Masucci L, Gasbarrini A, Tortora G, Cammarota G, Zitvogel L, Segata N, Iacovelli R, Ianiro G

Paper source

Open DOI

Αlpha¹ Rigor Review

Rigor Review Verified

Critical

High-priority

Action items to resolve · Critical / High

Confidence: highEngine: 7.6.0Source: full text

This Alpha1 Rigor Review uses AI Rigor Reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.

This is a well-conducted phase 2a RCT with strong scientific premise, adequate study design, and transparent reporting. The main weaknesses are incomplete software versioning and code sharing, a vague clinical data access statement, and the discovery of 4 inconsistently recomputed p-values (3 crossing significance thresholds) that require author clarification.

Three independent reviewer evaluations were synthesized; divergence on statistical analysis and data code availability was resolved by weighing verification component evidence. The statistics component recomputed 23 tests (19 consistent, 4 inconsistent with 3 decision errors); 72% of statistical claims remain unverified. No retracted or unfindable references were found.

12 major claims checked against the paper's own evidence: all adequately supported.

supportedResults, Primary and secondary efficacy outcomesReviewer 1
Donor FMT significantly improved median PFS compared with placebo FMT in patients with mRCC receiving pembrolizumab + axitinib.
The claim is supported by the reported median PFS values (24.0 vs 9.0 months, HR=0.50, P=0.035) with consistent findings in the per-protocol and post-hoc analyses.
Evidence: Results section: 'Median PFS was significantly improved in the d-FMT arm (24.0 months, 95% CI: 8.0–40.0 months) compared with the p-FMT arm (9.0 months, 95% CI: 2.2–15.2 months) (hazard ratio = 0.50, 90% CI: 0.27–0.92, P = 0.035; Fig. )'
“Median PFS was significantly improved in the d-FMT arm (24.0 months, 95% CI: 8.0–40.0 months) compared with the p-FMT arm (9.0 months, 95% CI: 2.2–15.2 months) (hazard ratio = 0.50, 90% CI: 0.27–0.92, P = 0.035; Fig. )”
supportedResults, Safety assessmentReviewer 1
Donor FMT was safe, with no FMT-related SAEs reported.
The safety analysis reports no deaths related to experimental treatments, no transmission of infectious agents, and rare procedure-related AEs (e.g., one grade 2 diarrhea). The safety set includes 49 patients.
Evidence: Results, Safety assessment: 'No deaths related to experimental treatments were reported. No transmission of any infectious agent after d-FMT was observed.'
“No deaths related to experimental treatments were reported. No transmission of any infectious agent after d-FMT was observed.”
supportedResults, Assessment of microbiome changes after treatmentsReviewer 1
Donor FMT resulted in increased α-diversity and larger microbiome shifts (β-diversity) compared with placebo.
The paper reports significant increases in Shannon α-diversity at multiple time points in the d-FMT arm but not in the p-FMT arm, and significantly higher Bray-Curtis dissimilarity from baseline at week 4 in the d-FMT arm.
Evidence: Results, Assessment of microbiome changes after treatments: 'we observed an increase in Shannon α-diversity versus baseline at week 1 (P = 0.05), week 4 (P < 0.001), week 12 (P = 0.02) and week 24 (P = 0.048) follow-ups ... significantly higher Bray–Curtis dissimilarity between posttreatment and baseline microbiomes in d-FMT versus p-FMT patients at the week 4 follow-up (P = 0.015)'
“we observed an increase in Shannon α-diversity versus baseline at week 1 (P = 0.05), week 4 (P < 0.001), week 12 (P = 0.02) and week 24 (P = 0.048) follow-ups ... significantly higher Bray–Curtis dissimilarity between posttreatment and baseline microbiomes in d-FMT versus p-FMT patients at the week 4 follow-up (P = 0.015)”
supportedResults, Quantification of donor microbiome engraftmentReviewer 1
Donor strain engraftment was higher in the d-FMT arm, confirming microbiological success of the FMT.
The DoSER metric shows significantly higher engraftment in the d-FMT arm (e.g., median 18.1% at week 1) compared with the p-FMT arm (medians ~1-2%), with P<0.001 across all timepoints.
Evidence: Results, Quantification of donor microbiome engraftment: 'The DoSER in patients receiving d-FMT was consistently higher than in the p-FMT arm throughout the whole study period (P < 0.001 across all timepoints starting from week 1 follow-up; Fig. )'
“The DoSER in patients receiving d-FMT was consistently higher than in the p-FMT arm throughout the whole study period (P < 0.001 across all timepoints starting from week 1 follow-up; Fig. )”
supportedResults, Donor microbiome engraftment effects on clinical outcomesReviewer 1
The absolute number of donor strains engrafting (DoSER) did not correlate with clinical benefit, but acquisition of specific donor strains (e.g., Blautia wexlerae) was positively associated with 12-month PFS.
The paper reports no significant difference in DoSER between PFS>12 and <12 subgroups. Strain-level analysis shows a significant positive association with Blautia wexlerae acquisition (50% vs 0%, P=0.047) and inverse associations with Oscillospiraceae SGB14845 and Akkermansia massiliensis.
Evidence: Results, Donor microbiome engraftment effects on clinical outcomes: 'the acquisition of the Blautia wexlerae (SGB4837) strain from the donor at week 1 was positively associated with 12-month PFS (percentage of recipients acquiring strain with versus without PFS > 12 months: 50% versus 0%, P = 0.047; Fig. )'
“the acquisition of the Blautia wexlerae (SGB4837) strain from the donor at week 1 was positively associated with 12-month PFS (percentage of recipients acquiring strain with versus without PFS > 12 months: 50% versus 0%, P = 0.047; Fig. )”
supportedResults, Post hoc analysesReviewer 1
Donor FMT had a stronger effect in patients with intermediate- or poor-prognosis disease (IMDC criteria).
Post-hoc analysis shows larger absolute differences and significant P values for 12-month PFS (63% vs 27%, P=0.045) in the intermediate/poor subgroup. The claim is appropriately qualified as exploratory.
Evidence: Results, Post hoc analyses: 'In this subpopulation, differences between the two treatment groups became larger in all evaluated efficacy outcomes. The median PFS was 18.8 months... 5.1 months (P = 0.033). The 12-month PFS was achieved by 10 of 16 patients (63%) in the d-FMT arm and by four of 15 patients (27%) in the p-FMT arm (P = 0.045).'
“In this subpopulation, differences between the two treatment groups became larger in all evaluated efficacy outcomes. The median PFS was 18.8 months... 5.1 months (P = 0.033). The 12-month PFS was achieved by 10 of 16 patients (63%) in the d-FMT arm and by four of 15 patients (27%) in the p-FMT arm (P = 0.045).”
supportedResults, Primary and secondary efficacy outcomesReviewer 2
Donor FMT significantly improved median PFS compared with placebo FMT.
The result is clearly reported with HR=0.50, 90% CI 0.27-0.92, P=0.035, and is supported by per-protocol sensitivity analysis (HR=0.43, P=0.028).
Evidence: Results: median PFS 24.0 vs 9.0 months, HR=0.50, P=0.035
“Median PFS was significantly improved in the d-FMT arm (24.0 months, 95% CI: 8.0–40.0 months) compared with the p-FMT arm (9.0 months, 95% CI: 2.2–15.2 months) (hazard ratio = 0.50, 90% CI: 0.27–0.92, P = 0.035)”
supportedResults, Primary and secondary efficacy outcomesReviewer 2
The primary endpoint (12-month PFS rate) was not met but showed a trend favoring d-FMT.
The result is transparently reported: 70% vs 41%, P=0.053, which did not reach the pre-specified significance threshold.
Evidence: Results: 12-month PFS d-FMT 70% vs p-FMT 41%, P=0.053
“the proportion of patients without progression or death 12 months after randomization was higher in the d-FMT arm than in the p-FMT arm (d-FMT: 16/23 patients, 70%; p-FMT: 9/22 patients, 41%; P = 0.053)”
supportedResults, Safety assessmentReviewer 2
Donor FMT is safe with no FMT-related SAEs.
Safety data are reported, including a single grade 2 diarrhea and one grade 3 oral mucositis in p-FMT, and no FMT-related deaths.
Evidence: Results, Safety assessment: 'No deaths related to experimental treatments were reported. No transmission of any infectious agent after d-FMT was observed.'
“No deaths related to experimental treatments were reported.”
supportedResults, Assessment of microbiome changes; Quantification of donor microbiome engraftmentReviewer 2
Donor FMT led to a microbiologically successful engraftment including increased alpha-diversity and donor strain engraftment.
The paper provides extensive evidence: increased Shannon diversity at multiple timepoints, higher species acquisition, higher Bray-Curtis dissimilarity, and DoSER consistently higher in d-FMT arm.
Evidence: Results, Assessment of microbiome changes and DoSER: increase in Shannon diversity, DoSER median 18.1% at week 1 vs ~2% in placebo.
“the donor FMT was microbiologically successful in our study.”
supportedResults, Donor microbiome engraftment effectsReviewer 2
Acquisition of Blautia wexlerae donor strain was positively associated with 12-month PFS.
The association is supported by a Fisher's exact test (P=0.047) with the specific strain acquisition data; however, the analysis is exploratory and in a small sample (N=13 for the week 1 d-FMT subgroup), so it is partially supported.
Evidence: Results, Donor microbiome engraftment effects: '50% versus 0%, P = 0.047'
“The acquisition of the Blautia wexlerae (SGB4837) strain from the donor at week 1 was positively associated with 12-month PFS (percentage of recipients acquiring strain with versus without PFS > 12 months: 50% versus 0%, P = 0.047)”
supportedResults, Donor microbiome engraftment effectsReviewer 2
Acquisition of A. massiliensis donor strain was inversely associated with 12-month PFS.
The inverse association is reported with P=0.006, a clear statistical result, though again in a small exploratory subgroup.
Evidence: Results: '0% versus 57%, P = 0.006'
“Acquisition of the donor strain of the rarer—as opposed to the prevalent A. muciniphila (SGB9226)— A. massiliensis sp. nov. (SGB9228) (percentage of recipients acquiring strain with versus without PFS > 12 months: 0% versus 57%, P = 0.006)”

Recomputed 22 tests: 19 consistent, 3 inconsistent (3 change significance at p<.05), 22 via agent-written checks. 1 reported summary statistic mathematically impossible for the stated N (GRIM/GRIMMER).

Rigor

Eight Dimensions7 pass · 1 warn

12 major claims12 supported

8 copyedit issues flagged: mostly consistency, grammar, clarity.

Checked 65 references: 64 verified — 1 not checked.

Reproducibility

5 data/code links checked; 5 live.

dataENALIVEHTTP 200
http://www.ebi.ac.uk/ena/data/view/PRJEB94043
Resolves to ENA (data repository).
datahttps://clinicaltrials.gov/study/NCT04758507LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://www.sealedenvelope.comLIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://clinicaltrials.gov/study/NCT04758507?term=ianiro&;rank=2#study-planLIVEHTTP 200
Resolves, but the content could not be matched to the paper.
codeGitHubLIVEHTTP 200
https://github.com/SegataLab/preprocessing
Resolves to GitHub (code repository).

Registered (1 ID: ClinicalTrials.gov). Reporting guideline cited: CONSORT.

Alpha¹ Rigor Action ItemsFinal pass before journal submission

Ready after substantive revision: the 4 inconsistent p-values (3 crossing significance thresholds) must be investigated and corrected or explained before submission. The remaining items (code sharing, software versioning, copyedit fixes) are manageable upgrades.

1.
HIGHstatistics
Investigate and correct the 4 inconsistently recomputed p-values in the statistical analysis: the reported p=0.167 and p=0.146 for HR 0.36 (95% CI 0.13–0.99) recomputed as p=0.049. Explain the discrepancy or correct the values in the Results and Abstract.
These discrepancies cross the significance threshold and could affect interpretation of secondary endpoints; an unexplained discrepancy undermines trust in all reported statistics.
2.
HIGHdata code
Deposit custom analysis code (strain-engraftment pipeline, strain replacement rate calculation) in a public repository (e.g., GitHub, Zenodo) with a persistent DOI, and reference the repository in the Data availability section.
Custom code is not shared, which prevents full reproduction of the microbiome analyses and violates best practices for computational reproducibility.
3.
HIGHdata code
Upgrade the patient-level clinical data availability statement by naming a managed-access platform (e.g., Vivli, YODA) or providing a clear data access committee contact with an expected response timeframe.
The current vague statement ('may be shared under a data use agreement') does not meet journal standards for a reproducible data availability plan.
4.
HIGHreporting
Provide version numbers for all major bioinformatics software components (e.g., HUMAnN, KneadData, all bioBakery suite tools) used in the microbiome analysis in the Methods section, not just in supplementary materials.
Incomplete software versioning undermines reproducibility of the bioinformatics pipeline.
5.
HIGHstatistics
Report exact p-values for all comparisons (e.g., ORR difference between arms, non-significant OS results) rather than only reporting confidence intervals or threshold statements.
Exact p-values allow readers to fully assess statistical significance; their absence is a reporting gap flagged by multiple reviewers.
6.
HIGHstatistics
Add a justification for the use of a one-sided 5% significance level in the sample size calculation, or provide a two-sided equivalent, in the Statistical analysis section.
One-sided tests are unusual for superiority trials and require explicit justification to avoid criticism during review.
7.
MEDIUMrigor
Pre-specify outlier handling criteria for both clinical and microbiome outcomes (e.g., criteria for excluding samples with very low sequencing depth, extreme diversity values) in the Methods or statistical analysis plan.
The absence of pre-specified outlier criteria leaves room for post-hoc decisions that can bias results.
8.
MEDIUMcopyedit
Fix incomplete figure references in the Results section: 'Fig. ' and 'Extended Data Fig. ' need the correct figure numbers inserted.
Missing figure references prevent readers from locating the corresponding data and could be seen as sloppy by reviewers.
9.
MEDIUMcopyedit
Correct the typo in Results paragraph 1: 'PFS > 12 months versus PFS > 12 months' should read 'PFS > 12 months versus PFS ≤ 12 months' or equivalent.
This is a clear copy-paste error that would confuse readers and could be misinterpreted as a logical inconsistency.
10.
MEDIUMcopyedit
Fix grammar in Abstract last sentence: change 'deserves further investigations' to 'warrant further investigation'.
Subject-verb agreement error that may be flagged by journal copyeditors.
11.
MEDIUMcopyedit
Remove the duplicate 'FMT' in Results, Assessment of microbiome changes: 'FMT had a considerable modulatory effect of FMT also on microbiome composition' should be rephrased.
Repetition of 'FMT' in the same sentence is a grammatical error that reduces clarity.
12.
MEDIUMcopyedit
Check Figure 4d and 4e descriptions in the legend: they are described with nearly identical text; confirm the intended differences between panels and correct if needed.
Potentially duplicated figure panels would misrepresent the data and confuse readers.
13.
LOWreporting
Add a brief statement in the Statistical analysis section describing how proportionality of hazards was checked (e.g., Schoenfeld residuals) rather than only stating it was checked.
Adding methodological detail strengthens the reader's confidence in the analysis without requiring additional analyses.
14.
LOWreporting
Consider adding a statement about planned independent replication or external validation of key findings (e.g., the strain-level associations with PFS) as a future direction in the Discussion.
Acknowledging the need for replication would strengthen the paper's credibility and is expected for phase 2a trials.

Full text (86,308 chars)Analyzed: Jul 7, 2026

Αlpha¹ assesses methodological rigor, not the importance of the findings. See how we calibrate →