# Neoadjuvant immunotherapy and chemotherapy regimens for the treatment of high-risk, early-stage triple-negative breast cancer: a systematic review and network meta-analysis.

- **DOI:** [10.1186/s12885-023-11293-4](https://doi.org/10.1186/s12885-023-11293-4)
- **Rating:** Sound (3.3 out of 5)
- **Verdict:** A useful comparative TNBC treatment synthesis for HTA decisions, constrained by a sparse indirect network.
- **Framework version:** v5.5
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-07-25
- **Evaluation page:** https://www.nabu.science/paper/10.1186%2Fs12885-023-11293-4

## Key claims

1. Neoadjuvant pembrolizumab + chemotherapy followed by adjuvant pembrolizumab confers benefits in response and survival outcomes versus alternative neoadjuvant treatments for early-stage TNBC.
   - Support: Tentative
   - Type: Primary finding
   - Basis: broad conclusion rests on sparse, mostly indirect comparisons with favorable results limited by comparator and outcome
   - Location: Abstract, Conclusions
2. The relative efficacy of neoadjuvant pembrolizumab + chemotherapy followed by adjuvant pembrolizumab was favorable to paclitaxel followed by anthracycline + cyclophosphamide in terms of pCR, EFS, and OS.
   - Support: Supported
   - Type: Secondary
   - Basis: reported NMA estimates are statistically favorable across all three named outcomes
   - Location: Abstract, Results; Tables 3–5
3. The relative efficacy of neoadjuvant pembrolizumab + chemotherapy followed by adjuvant pembrolizumab was favorable to paclitaxel + carboplatin followed by anthracycline + cyclophosphamide in terms of pCR and EFS.
   - Support: Supported
   - Type: Secondary
   - Basis: specific pCR and EFS estimates are favorable, though OS does not meet the same threshold
   - Location: Abstract, Results; Tables 3–4
4. The relative efficacy of neoadjuvant pembrolizumab + chemotherapy followed by adjuvant pembrolizumab was favorable to paclitaxel + bevacizumab followed by anthracycline + cyclophosphamide + bevacizumab in terms of pCR.
   - Support: Supported
   - Type: Secondary
   - Basis: claim is outcome-specific and matches the statistically favorable pCR comparison
   - Location: Abstract, Results; Table 3
5. The relative efficacy of neoadjuvant pembrolizumab + chemotherapy followed by adjuvant pembrolizumab was favorable to paclitaxel + carboplatin + veliparib followed by anthracycline + cyclophosphamide in terms of EFS.
   - Support: Supported
   - Type: Secondary
   - Basis: claim is limited to EFS and aligns with the reported credible interval
   - Location: Abstract, Results; Table 4

## Strengths

- **Clinically focused comparative evidence gap** The study targets the absence of head-to-head comparisons between the approved pembrolizumab regimen and other neoadjuvant treatments for early-stage TNBC. This creates a clear clinical and policy use case for an NMA.
  - Dimension: contribution
  - Location: Background
- **Broad search and duplicate review process** The methods report searches of EMBASE, MEDLINE, CENTRAL, conference proceedings, and clinical trial registries, followed by duplicate abstract selection, full-text selection, and data extraction. These procedures support the completeness and reproducibility of the review process.
  - Dimension: methodological rigour
  - Location: Methods, Systematic Review
- **Transparent presentation of comparative estimates** The article presents full matrices for pCR, EFS, and OS, including statistically meaningful and non-meaningful comparisons. This lets readers see that favorable findings differ by comparator and outcome rather than applying uniformly.
  - Dimension: reporting
  - Location: Results, Tables 3–5
- **Direct pathway to HTA use** The background explicitly frames NMA as evidence for clinicians, guideline developers, and health technology assessment agencies. The review’s stated purpose aligns with a concrete decision-support pathway.
  - Dimension: application pathway
  - Location: Background; Methods, Systematic Review

## Limitations

- **Sparse network limits comparative confidence** Only one trial connected each treatment in the network, so between-study heterogeneity could not be reliably estimated and a fixed-effects assumption was used. This is a central limitation for the indirect comparisons.
  - Dimension: methodological rigour
  - Location: Methods, Feasibility Assessment and Network Meta-analysis
- **Post-review comparator broadening introduced flexibility** After completion of the systematic review, the comparator component of the PICOS was broadened to form a connected network. That change may be defensible for feasibility, but it reduces confidence that the evidence assembly was fully prespecified.
  - Dimension: methodological rigour
  - Location: Methods, Systematic Review
- **Transitivity support is incompletely documented** The feasibility assessment notes that baseline patient characteristics were not reported for most trials, yet the NMA depends on assumptions of sufficient similarity across trials. The paper also assumes PD-L1 expression influenced only the pembrolizumab-containing regimen.
  - Dimension: methodological rigour
  - Location: Results, Feasibility Assessment; Results, Network Meta-analysis
- **Conclusions are broader than results** The conclusion states that the pembrolizumab regimen is effective compared with other neoadjuvant treatments, while the results show statistically favorable findings only for specific comparators and outcomes. This creates an internal-coherence watch-out rather than a fatal inconsistency.
  - Dimension: contribution
  - Location: Conclusions; Results, Tables 3–5

## Reliability checks

Caveats (4 of 4 checks).

- **Internal coherence:** Noted. The article is broadly internally consistent, with results tables aligning with the abstract’s named favorable comparisons. Interpretive watch-outs remain because the network was broadened after the initial review and the conclusions generalize beyond some comparator- and outcome-specific findings.
- **Research conduct:** Clean. No text-grounded conduct problem meets the threshold for a reliability concern. The protocol was not registered, but this is explicitly disclosed and explained as related to health technology assessment submissions.
- **Reference integrity:** Clean. 35 references verified via PubMed, Crossref, or OpenAlex.
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
