Evidence grades (GRADE-adapted): A high — multiple well-conducted RCTs or systematic reviews converge. B moderate — single pivotal RCT or consistent observational evidence. C limited — single observational study, mechanistic, or expert consensus. D preclinical / hypothesis-generating.
Why residual disease matters
Patients with residual disease after neoadjuvant chemotherapy have substantially higher recurrence and mortality risks than patients who achieve pathologic complete response. In the CTNeoBC pooled analysis (see the pCR endpoint synthesis), the TNBC pCR-vs-no-pCR EFS HR was 0.24 — meaning no-pCR TNBC patients had approximately four-fold higher event rates than pCR achievers. The Symmans 2017 long-term analysis quantified this further: 10-year recurrence-free survival was approximately 90% for pCR/RCB-I and approximately 30–40% for RCB-III.
The clinical implication: residual-disease patients need additional adjuvant therapy beyond what they received neoadjuvantly. The question is which agents, in what combination, and for how long.
Adjuvant capecitabine — the CREATE-X precedent
Masuda and colleagues, on behalf of the Japanese Breast Cancer Study Group, randomized 910 patients with HER2-negative residual breast cancer after standard neoadjuvant chemotherapy to adjuvant capecitabine (1,250 mg/m² twice daily, 8 cycles, approximately 6 months) vs observation[1]A. The trial enrolled both HR+ and HR− patients; the TNBC subgroup was 286 patients.
Headline results
- Whole-trial 5-year DFS: 74.1% vs 67.6% (HR 0.70; 95% CI 0.53–0.92)
- Whole-trial 5-year OS: 89.2% vs 83.6% (HR 0.59; 95% CI 0.39–0.90)
- TNBC subgroup 5-year DFS: 69.8% vs 56.1% — substantial absolute benefit
- TNBC subgroup 5-year OS: 78.8% vs 70.3%
The CREATE-X result established adjuvant capecitabine for 6 months as standard of care in residual TNBC after standard pre-IO-era neoadjuvant chemotherapy. NCCN guidelines incorporated this in 2017.
CREATE-X in the post-KEYNOTE-522 era
A persistent uncertainty: the CREATE-X data preceded the KEYNOTE-522 regimen (pembrolizumab + carboplatin + paclitaxel → pembrolizumab + AC). Whether adjuvant capecitabine retains efficacy after the more intensive KEYNOTE-522 regimen, which already includes carboplatin and pembrolizumab, is not directly tested by CREATE-X. In current clinical practice, capecitabine is commonly added to adjuvant pembrolizumab continuation for residual-disease TNBC patients, but the evidence base for this practice is precedent-based rather than randomized[2]B.
The capecitabine toxicity profile (hand-foot syndrome, diarrhea, fatigue, modest myelosuppression) is generally manageable; serious toxicities are uncommon. Hand-foot prophylaxis with topical urea-based creams reduces the most common patient-quality-of-life issue.
Adjuvant olaparib — OlympiA in germline-BRCA-mutated disease
Tutt and colleagues randomized 1,836 patients with germline BRCA1/2 pathogenic variants and high-risk early-stage breast cancer (residual disease after neoadjuvant chemotherapy, OR adjuvant-setting high-risk features) to olaparib 300 mg twice daily for 1 year vs placebo[3]A. The TNBC subgroup was approximately 80% of enrollment (~1,470 patients).
Headline results
- 3-year invasive disease-free survival (primary endpoint): 85.9% vs 77.1% (HR 0.58; 95% CI 0.46–0.74)
- 3-year distant DFS: 87.5% vs 80.4% (HR 0.57)
- 4-year OS (2022 update): 89.8% vs 86.4% (HR 0.68; p=0.009)
- Safety profile: hematologic and GI toxicities at expected rates; no excess MDS/AML at 3.5-year follow-up
OlympiA established adjuvant olaparib for 1 year as standard of care in germline-BRCA-mutated high-risk early-stage breast cancer. FDA approval came in March 2022. ESMO and NCCN guidelines have incorporated.
OlympiA + KEYNOTE-522 + capecitabine combination
A germline-BRCA-positive TNBC patient with residual disease after the KEYNOTE-522 regimen now has three concurrent adjuvant options on the table:
- Adjuvant pembrolizumab continuation (per the KEYNOTE-522 protocol)
- Adjuvant capecitabine (CREATE-X precedent)
- Adjuvant olaparib 1 year (OlympiA)
All three concurrently is not formally tested in a randomized trial but is increasingly common in clinical practice. The toxicity profiles are sufficiently non-overlapping to make the combination tolerated; the three together produce a substantial pill burden and side-effect mix that requires close monitoring and supportive care. Whether all three are needed, or whether the marginal benefit of capecitabine is preserved when olaparib and pembrolizumab are already given, is unresolved.
Adjuvant pembrolizumab continuation
The KEYNOTE-522 protocol prescribes pembrolizumab continuation for approximately 9 months after surgery (an additional 9 doses, every 3 weeks), regardless of pCR status. The EFS benefit observed in KEYNOTE-522's 2022 update[4]A was driven primarily by the residual-disease subgroup; pCR achievers had near-equivalent EFS in both arms.
This pattern motivates the de-escalation question: do pCR achievers actually need the adjuvant pembrolizumab continuation, or could they stop after surgery? The OptimICE-pCR trial is randomizing pCR achievers to continue vs observation; primary readout expected in the next 2–3 years.
For residual-disease patients, the adjuvant pembrolizumab continuation is currently standard. The combination with adjuvant capecitabine and (in BRCA+) adjuvant olaparib is operationally complex but increasingly common.
Emerging adjuvant ADC trials
Several phase III trials are testing whether antibody-drug conjugates can further improve outcomes in residual-disease TNBC patients:
- SASCIA (NCT04595565) — adjuvant sacituzumab govitecan vs treatment of physician's choice in residual disease after standard neoadjuvant chemotherapy. Enrolling.
- TROPICS-08 (NCT04595565) — adjuvant sacituzumab govitecan in high-risk residual disease. Enrolling.
- ASCENT-05 (NCT05633654) — adjuvant sacituzumab govitecan + pembrolizumab vs pembrolizumab + treatment of physician's choice. Residual disease after KEYNOTE-522 regimen specifically. Enrolling.
- DESTINY-Breast05 (NCT04622319) — adjuvant trastuzumab deruxtecan in HER2-positive residual disease (not TNBC but adjacent program). Pending readout.
If any of these trials are positive, the adjuvant residual-disease landscape for TNBC becomes substantially more complex with multiple ADC options to choose among. The sequencing / combination decisions will require considerable clinical judgment.
Practical adjuvant algorithm for residual-disease TNBC (2026 consensus)
Roughly the current US/EU practice for a residual-disease TNBC patient following the KEYNOTE-522 regimen:
- Confirm germline BRCA1/2 status — if not yet known. PALB2 testing increasingly considered.
- Continue adjuvant pembrolizumab per the KEYNOTE-522 protocol (9 additional doses).
- Add adjuvant capecitabine — ~6 months of treatment. Standard offering; CREATE-X precedent.
- If germline BRCA1/2 positive: add adjuvant olaparib 1 year, starting after completion of acute toxicity recovery.
- Consider clinical trial enrollment in SASCIA, ASCENT-05, or similar adjuvant-ADC trials.
- Standard surgical + radiation completion per breast cancer-general principles.
Evidence table
| Trial / Agent | Setting | Population | Headline outcome |
|---|---|---|---|
| CREATE-X (capecitabine) | Adjuvant; residual disease after standard neoadj chemo | HER2−, TNBC subgroup ~286 | 5-yr DFS 69.8% vs 56.1% (TNBC); OS 78.8% vs 70.3% |
| OlympiA (olaparib) | Adjuvant; germline BRCA1/2 + high risk | n=1,836, ~80% TNBC | 3-yr iDFS HR 0.58; 4-yr OS HR 0.68 |
| KEYNOTE-522 adjuvant phase | Adjuvant pembro continuation | Stage II–III TNBC, post-neoadj | 3-yr EFS HR 0.63 (benefit concentrated in residual-disease subset) |
| SASCIA (sacituzumab) | Adjuvant; residual after neoadj | ~1,500 enrollment target | Pending |
| ASCENT-05 (sacituzumab + pembro) | Adjuvant after KEYNOTE-522 | Residual disease, all-comers | Pending |
Open questions and active investigation
- Is the capecitabine benefit preserved after KEYNOTE-522? CREATE-X predates the IO regimen. Whether adjuvant capecitabine adds incremental benefit to adjuvant pembrolizumab continuation in residual-disease patients is not formally answered. Current consensus practice is yes, by precedent; randomized confirmation is awaited.
- Sequencing of multiple adjuvant agents. Pembrolizumab continuation, capecitabine, and (in BRCA+) olaparib together is operationally complex. Whether they should overlap or be staggered, and whether the full duration of each is needed, hasn't been formally tested.
- Adjuvant ADC trial readouts. SASCIA, TROPICS-08, ASCENT-05 results will determine whether adjuvant ADCs become standard. If positive, the residual-disease landscape becomes substantially more complex.
- De-escalation in pCR achievers. OptimICE-pCR and other trials testing reduced adjuvant pembrolizumab in pCR achievers will define whether the full 9-month adjuvant phase is needed.
- Biomarker-guided adjuvant selection. RCB, ctDNA MRD status, tumor genomic features (BRCA mutation, HER2-low, TIL density at residual disease) might guide which adjuvant agents to use. Currently not formalized in clinical practice but actively investigated.
- Duration of adjuvant olaparib. OlympiA used 1-year duration. Whether shorter or longer would be equivalent or superior is not directly tested.
For the neoadjuvant regimen that precedes adjuvant decision-making, see the KEYNOTE-522 synthesis. For pCR endpoint detail, see the pCR as endpoint synthesis. For PARP inhibitor biology, see the PARP synthesis. For the patient-layer companion, see Treatment options.
References
Each citation links to the original publication via DOI. The same records are searchable in the evidence library by title or DOI.
- Masuda N, Lee SJ, Ohtani S, et al. Adjuvant Capecitabine for Breast Cancer after Preoperative Chemotherapy (CREATE-X). N Engl J Med. 2017;376(22):2147–2159. doi:10.1056/NEJMoa1612645. ↩
- Loibl S, Poortmans P, Morrow M, Denkert C, Curigliano G. Breast cancer. Lancet. 2021;397(10286):1750–1769. doi:10.1016/S0140-6736(20)32381-3. ↩
- Tutt ANJ, Garber JE, Kaufman B, et al. Adjuvant Olaparib for Patients with BRCA1- or BRCA2-Mutated Breast Cancer (OlympiA). N Engl J Med. 2021;384(25):2394–2405. doi:10.1056/NEJMoa2105215. ↩
- Schmid P, Cortes J, Dent R, et al. Event-free Survival with Pembrolizumab in Early Triple-Negative Breast Cancer (KEYNOTE-522). N Engl J Med. 2022;386(6):556–567. doi:10.1056/NEJMoa2112651. ↩
Last reviewed: 2026-06-04. Researcher-layer synthesis page. Evidence grades follow the GRADE-adapted rubric defined at the top of this page.