Highlight
– In a randomized phase III trial of 717 patients with triple‑negative breast cancer (TNBC), adding weekly carboplatin (AUC‑2) to a sequential paclitaxel → anthracycline‑cyclophosphamide neoadjuvant program did not meet the primary end point of event‑free survival (EFS) (HR 0.80; 95% CI 0.62–1.03; P = .081).
– A secondary analysis showed improved overall survival (OS) with carboplatin (HR 0.74; 95% CI 0.57–0.97; nominal P = .029), and a strong, statistically significant interaction by menopausal status: benefit concentrated in premenopausal patients (EFS HR 0.61; OS HR 0.57), with no apparent benefit in postmenopausal patients.
– Carboplatin increased grade ≥3 myelosuppression without increased nonhematologic toxicity. Clinical application should weigh modest survival signals against toxicity, reproductive considerations, and evolving integration with immunotherapy and biomarker‑guided strategies.
Background: Clinical context and unmet need
Triple‑negative breast cancer (TNBC) accounts for roughly 10–20% of breast cancers and is characterized by absence of estrogen receptor (ER), progesterone receptor (PR) and HER2. TNBC is clinically heterogeneous but tends to be more aggressive, has higher early recurrence risk, and lacks targeted endocrine or anti‑HER2 options, making cytotoxic chemotherapy the backbone of curative‑intent management.
Neoadjuvant chemotherapy is commonly used in stage II–III TNBC because it can downstage tumors, allow breast‑conserving surgery, and provide in vivo assessment of chemosensitivity — notably pathologic complete response (pCR), which is associated with improved long‑term outcomes. Multiple prior studies have examined whether adding platinum agents (typically carboplatin) improves pCR rates and survival in TNBC. Earlier randomized and nonrandomized data suggested improved pCR with carboplatin but inconsistent effects on longer‑term outcomes, and toxicity (notably hematologic adverse events) has been a concern.
Study design
Gupta and colleagues report a randomized, phase III trial enrolling 720 patients with TNBC between April 2010 and January 2020; 717 patients were included in the modified intention‑to‑treat analysis (carboplatin arm, n=361; control arm, n=356). Patients were stratified by clinical stage and menopausal status and received neoadjuvant paclitaxel (100 mg/m2 weekly) for 8 weeks followed by four cycles of anthracycline plus cyclophosphamide. The experimental arm received once‑weekly carboplatin (AUC‑2) concomitantly with paclitaxel; the control arm received the same regimen without carboplatin.
The prespecified primary end point was event‑free survival (EFS). Secondary end points included overall survival (OS) and pathologic complete response (pCR). The present report constitutes the prespecified primary analysis with median follow‑up of 67.6 months.
Key findings
Primary outcome — event‑free survival
At a median follow‑up of 67.6 months, there were 111 EFS events in the carboplatin arm versus 131 events in the control arm (HR for EFS 0.80; 95% CI 0.62–1.03; two‑sided unstratified P = .081). Estimated 5‑year EFS was 70.7% (95% CI 65.8%–75.6%) in the carboplatin arm versus 64.1% (95% CI 59.0%–69.2%) in the control arm.
Interpretation: The study did not achieve its primary end point at conventional significance thresholds. The point estimate (HR 0.80) numerically favors carboplatin and the absolute 5‑year EFS difference (~6.6%) could be clinically meaningful, but uncertainty remains because the confidence interval includes unity and the prespecified P threshold was not met.
Secondary outcome — overall survival
There were 94 deaths in the carboplatin arm and 121 deaths in the control arm (HR for OS 0.74; 95% CI 0.57–0.97; nominal P = .029). Estimated 5‑year OS was 74.4% with carboplatin versus 66.8% without carboplatin.
Interpretation: OS was improved with carboplatin by a relative 26% reduction in the hazard of death. However, OS was a secondary end point and the reported P value is nominal; multiplicity adjustments for multiple testing were not detailed in the summary, so the OS signal should be interpreted cautiously and ideally confirmed in independent data.
Subgroup analysis — menopausal status interaction
The authors report a statistically significant interaction by menopausal status: premenopausal patients derived substantial benefit from carboplatin (EFS HR 0.61; 95% CI 0.43–0.84; nominal P = .003; 5‑year EFS 75.0% vs 59.6%; OS HR 0.57; 95% CI 0.40–0.82; nominal P = .002; 5‑year OS 78.2% vs 64.6%). In contrast, postmenopausal patients experienced no benefit (EFS HR 1.19; 95% CI 0.80–1.78; OS HR 1.06; 95% CI 0.70–1.61).
Interpretation: A statistically significant interaction suggests effect modification by menopausal status. Potential explanations include differences in underlying tumor biology (higher prevalence of germline BRCA mutations or homologous recombination deficiency in younger patients), differences in chemotherapy tolerance or intensity, or chance findings. Because subgroup analyses can be prone to false positives, these findings should prompt further investigation and validation rather than immediate, universal practice change.
Pathologic complete response and toxicity
The report lists pCR as a secondary end point; the summary provided does not include the pCR rates or their relation to long‑term outcomes, which are important for mechanistic interpretation. Prior trials have shown that carboplatin tends to increase pCR rates in TNBC. This trial found more grade ≥3 myelosuppression in the carboplatin arm; nonhematologic toxicities were similar across arms. Increased hematologic toxicity is a predictable and clinically important tradeoff that requires proactive management (growth factor support, dose modifications, transfusion where necessary).
Expert commentary: contextualizing the results
How do these findings fit with prior evidence? Earlier randomized neoadjuvant trials — notably GeparSixto and CALGB (Alliance) 40603 — demonstrated improved pCR with carboplatin, but durable survival benefits were inconsistent. GeparSixto suggested improvements in disease‑free survival in a subset analysis; CALGB 40603 showed increased pCR but no significant EFS/OS advantage. Recent advances complicate interpretation: the KEYNOTE‑522 trial established that adding pembrolizumab (an anti‑PD‑1 antibody) to neoadjuvant chemotherapy increased pCR and EFS for early TNBC, and immunotherapy is now incorporated into some neoadjuvant regimens. The present trial spans 2010–2020, a period before routine integration of checkpoint inhibitors into early TNBC care, limiting direct applicability to current standard regimens that may already include immunotherapy.
Possible biological rationale for premenopausal benefit: younger patients more often harbor deleterious germline BRCA1/2 mutations and other homologous recombination deficiency (HRD) features that sensitize tumors to DNA‑damaging agents such as platinum compounds. Ovarian hormonal milieu might conceivably influence tumor biology and chemosensitivity, although this is speculative. Confirmatory biomarker analyses (BRCA, HRD scores, genomic scarring) are needed to clarify which patients derive the most benefit.
Statistical considerations: the trial’s primary end point (EFS) was not met; the OS improvement is encouraging but secondary. When a trial misses its primary end point, emphasis on secondary end points requires caution because of potential type I error and multiplicity. The menopausal interaction is hypothesis‑generating and should be validated prospectively.
Clinical implications and practical guidance
For practicing clinicians considering carboplatin in the neoadjuvant management of TNBC, this study provides important, but not definitive, information:
- Premenopausal patients may derive a meaningful long‑term benefit from adding carboplatin, based on subgroup data. Discuss the potential OS advantage and the increased risk of hematologic toxicity during shared decision‑making.
- For postmenopausal patients, the trial found no benefit and therefore routine addition of carboplatin is not supported by this evidence.
- Integrate biomarker assessment (germline BRCA testing; if available, HRD testing) into preoperative evaluation as these may identify patients most likely to benefit from DNA‑damaging agents or PARP inhibitors.
- When carboplatin is used, anticipate and manage grade ≥3 myelosuppression with monitoring, dose adjustments, and supportive care (growth factors, transfusions) as appropriate. Counsel premenopausal patients about fertility preservation before cytotoxic chemotherapy and discuss gonadotropin‑releasing hormone agonists if indicated.
- Consider how carboplatin should be combined with immunotherapy: KEYNOTE‑522 and other immune‑chemotherapy data may alter the balance of risks and benefits, and prospective trials are needed to define optimal combinations and sequences.
Limitations and unanswered questions
Key limitations include the primary end point not being met, the secondary nature of the OS result (with nominal P values), lack of detailed pCR data in the provided summary, and changes in standard of care over the long recruitment period (2010–2020). The interplay with checkpoint inhibitors and contemporary standard regimens remains unresolved. Additionally, comprehensive biomarker analyses (BRCA status, HRD, PD‑L1) were not presented in the summary but are critical to personalize platinum use.
Conclusion and next steps
The trial by Gupta et al. provides important randomized evidence that adding weekly carboplatin to sequential taxane‑anthracycline neoadjuvant chemotherapy does not significantly improve EFS in an unselected TNBC population but is associated with improved OS and a clear benefit in premenopausal patients. These findings are hypothesis‑generating and suggest that younger patients — perhaps enriched for HRD/BRCA mutations — may benefit from platinum‑containing neoadjuvant regimens. Clinicians should weigh potential survival benefit against increased hematologic toxicity, and consider biomarker testing and fertility preservation in premenopausal patients.
Prospective validation and biomarker‑driven strategies are priorities. Future trials should clarify whether carboplatin remains beneficial when combined with modern immunotherapy‑containing neoadjuvant regimens and should define genomic predictors of benefit to enable precision use of platinum agents in early TNBC.
Funding and trial registration
For funding sources, trial registration, and detailed protocol specifics, refer to the original publication: Gupta S et al., J Clin Oncol. 2025 Oct 20: JCO-25-01023. DOI: 10.1200/JCO-25-01023.
Selected references
– Gupta S, Nair N, Hawaldar RW, et al. Addition of Carboplatin to Sequential Taxane-Anthracycline Neoadjuvant Chemotherapy in Triple‑Negative Breast Cancer: A Phase III Randomized Controlled Trial. J Clin Oncol. 2025 Oct 20. doi:10.1200/JCO-25-01023.
– von Minckwitz G, Schneeweiss A, Loibl S, et al. Neoadjuvant carboplatin in patients with triple‑negative and HER2‑positive early breast cancer (GeparSixto): a randomised phase 2 trial. Lancet Oncol. 2014;15(7):747‑756.
– Sikov WM, Berry DA, Perou CM, et al. Impact of the addition of carboplatin and/or bevacizumab to neoadjuvant chemotherapy on pathologic complete response rates in triple‑negative breast cancer: CALGB 40603 (Alliance). J Clin Oncol. 2015;33(1):13‑21.
– Schmid P, Cortes J, Pusztai L, et al. Pembrolizumab for early triple‑negative breast cancer. N Engl J Med. 2020;382:810‑821. (KEYNOTE‑522)
AI thumbnail prompt
Multidisciplinary oncology team in a modern clinic reviewing a patient’s neoadjuvant treatment plan on a tablet; foreground shows a stylized DNA double helix and a chemical formula of carboplatin overlaid faintly; muted clinical color palette, focused professional expressions, subtle hospital background, realistic style.
