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.
The combination rationale
The established pembrolizumab + chemotherapy regimens (KEYNOTE-355 metastatic; KEYNOTE-522 early-stage) demonstrate that chemotherapy contributes substantially to IO benefit in TNBC. The mechanistic understanding: chemotherapy produces immunogenic cell death, releasing tumor antigens, recruiting antigen-presenting cells, and priming T-cell responses that ICI can then sustain. This synergy is well-supported.
Whether other classes of therapy — PARP inhibitors, ADCs, radiation, or additional checkpoint blockade — produce analogous synergy with ICI is the open question driving multiple investigational programs. The biological rationale varies by combination class.
ICI + PARP inhibitor
Rationale: PARP inhibition produces unrepaired DNA damage in HR-deficient cells, increasing tumor mutational burden, neoantigen presentation, and activation of cGAS-STING innate immunity. These changes should make tumors more immunogenic and therefore more susceptible to ICI[1]B.
MEDIOLA (durvalumab + olaparib)
Domchek and colleagues conducted a phase II basket trial of durvalumab + olaparib in germline-BRCA-mutated metastatic breast cancer and other tumor types[2]B. The breast cancer cohort (n=34) showed:
- Disease control rate at 12 weeks: 80%
- Objective response rate: 63%
- Median PFS: 8.2 months
Promising single-arm data but without randomized comparison to olaparib monotherapy or to chemotherapy.
KEYLYNK-009 (pembrolizumab + olaparib)
Larger randomized testing of the combination in metastatic TNBC, with pembrolizumab + olaparib vs pembrolizumab + chemotherapy as the comparator. Enrollment included both germline-BRCA-mutated and unselected populations.
Initial reports suggested the combination did not exceed pembrolizumab + chemotherapy benefit, raising questions about whether PARP inhibition + ICI delivers true synergy or whether the chemotherapy backbone in standard pembrolizumab + chemo is providing most of the immunomodulatory benefit. Mature randomized data may refine this assessment.
ICI + ADC
Rationale: ADC payloads (SN-38 for sacituzumab; DXd for trastuzumab deruxtecan) induce DNA damage, microtubule disruption, or other cytotoxic effects that can produce immunogenic cell death similar to traditional chemotherapy. Combined with ICI, ADCs might produce greater immune-mediated benefit than either alone.
ASCENT-04 (sacituzumab + pembrolizumab)
Tests sacituzumab govitecan + pembrolizumab vs investigator's-choice chemotherapy + pembrolizumab in first-line metastatic TNBC with PD-L1 CPS ≥ 10. The trial design directly tests whether sacituzumab is a better IO partner than standard chemotherapy in the IO-eligible patient population.
BEGONIA (durvalumab + various combinations)
A phase I/II umbrella trial testing multiple ICI + ADC combinations including durvalumab + datopotamab deruxtecan (Dato-DXd) in first-line metastatic TNBC. Preliminary data suggest promising response rates; randomized data are pending[3]C.
Other ICI + ADC trials
- ICARUS-Breast01 — patritumab deruxtecan + immunotherapy in HER3-expressing TNBC
- Multiple investigator-initiated combinations testing ICI + various ADCs in pre-treated metastatic TNBC
ICI + radiation
Rationale: Radiation produces DNA damage and antigen release within the irradiated tumor; in combination with systemic ICI, this may produce "abscopal" effects (immune-mediated tumor regression at non-irradiated sites). The preclinical evidence for radiation + ICI synergy is substantial; clinical translation has been less consistent.
In metastatic TNBC, several trials are testing radiation + pembrolizumab combinations, including:
- Stereotactic body radiation therapy + pembrolizumab in oligometastatic TNBC
- Whole-brain or stereotactic radiation to brain metastases + pembrolizumab for CNS-only progression
- Radiation to single index lesions + concurrent IO + chemotherapy in metastatic disease
No randomized phase III data yet support routine clinical use of radiation + ICI combinations in TNBC beyond their standard individual indications.
Dual ICI strategies — TIGIT, LAG-3
Rationale: Tumors that fail single-agent PD-1/PD-L1 blockade may have additional immune-checkpoint pathways driving T-cell dysfunction. Targeting multiple checkpoints simultaneously is biologically plausible and has been successful in some cancer types (e.g., nivolumab + ipilimumab in melanoma; relatlimab + nivolumab in melanoma).
TIGIT inhibitors
TIGIT is a co-inhibitory receptor expressed on activated T cells and regulatory T cells; its blockade may augment PD-1 blockade. SKYSCRAPER trials testing tiragolumab + atezolizumab + chemotherapy in TNBC are ongoing. Early data in other cancer types (NSCLC) showed promising signals that did not consistently replicate in confirmatory trials.
LAG-3 inhibitors
LAG-3 is another co-inhibitory checkpoint; its blockade combined with PD-1 inhibition has FDA-approved indication in metastatic melanoma (relatlimab + nivolumab). Testing in TNBC is at earlier phase; preliminary signals in PD-L1+ subsets are exploratory.
Other dual-checkpoint combinations
- CTLA-4 + PD-1 (nivolumab + ipilimumab) in TNBC has been tested with limited single-agent activity and tolerability concerns
- TIM-3 + PD-1 combinations under investigation
- OX40 agonist + PD-1 inhibitor combinations in early-phase trials
ICI + targeted-therapy combinations
Additional combination strategies being investigated:
- ICI + AKT inhibitor — capivasertib + chemo + ICI testing in metastatic TNBC; PI3K pathway modulation hypothesized to alter immune microenvironment
- ICI + CDK4/6 inhibitor — palbociclib or abemaciclib + pembrolizumab; CDK4/6 inhibition increases tumor antigen presentation via cell-cycle arrest
- ICI + cancer vaccine — neoantigen vaccines combined with ICI to enhance tumor-specific T-cell priming
- ICI + bispecific antibodies — emerging class targeting two tumor antigens or one tumor antigen + one immune checkpoint
Most of these are at phase I/II proof-of-concept stage.
Evidence table
| Combination | Rationale | Trial | Status |
|---|---|---|---|
| Pembrolizumab + chemo | Established | KEYNOTE-355, KEYNOTE-522 | Approved standard |
| Durvalumab + olaparib | DDR + IO synergy | MEDIOLA | Promising; randomized comparison pending |
| Pembrolizumab + olaparib | DDR + IO synergy | KEYLYNK-009 | Did not outperform pembro + chemo |
| Sacituzumab + pembrolizumab | ADC + IO synergy | ASCENT-04 | Ongoing |
| Datopotamab deruxtecan + durvalumab | ADC + IO synergy | BEGONIA | Early signal positive |
| SBRT + pembrolizumab | Abscopal + IO | Multiple phase II | Hypothesis-generating |
| Tiragolumab + atezolizumab + chemo | Dual checkpoint | SKYSCRAPER series | Mixed signals across tumor types |
Open questions and active investigation
- Will any combination exceed pembrolizumab + chemo benefit? The KEYLYNK-009 result (PARP inhibitor combination did not exceed chemo combination) raises the possibility that chemotherapy is the most potent IO partner in TNBC. If true, the value-add from PARP, ADC, or dual checkpoint combinations may be modest.
- Biomarker selection for combinations. PD-L1 CPS gates pembrolizumab + chemo; whether different combinations require different biomarkers (HRD signature for PARP combinations, Trop-2 expression for sacituzumab combinations, HER2 expression for T-DXd combinations) is being studied.
- Toxicity profiles. Combination strategies multiply toxicity risks. ICI + PARP can produce overlapping fatigue and cytopenias; ICI + ADC can produce overlapping nausea and infusion reactions; ICI + radiation can produce overlapping skin reactions and pneumonitis risk. Whether the efficacy gains justify the toxicity costs is combination-specific.
- Sequencing vs combining. Some combination strategies may produce similar long-term outcomes via sequential use rather than concurrent combination, with lower toxicity. The randomized comparisons would inform sequencing-vs-combination preferences.
- Earlier-line combinations. Most IO combinations are tested in metastatic settings. Whether moving them earlier (first-line metastatic, neoadjuvant) would change the calculus is being investigated.
- Mechanism-of-action validation. Many combination rationales (immunogenic cell death, antigen release, microenvironment modulation) are hypothesized from preclinical work. Translational studies in trial patients with serial biopsies would test whether the hypothesized mechanism is engaged.
For the established pembrolizumab + chemo evidence base, see the IMpassion/KEYNOTE umbrella, the first-line metastatic synthesis, and the KEYNOTE-522 synthesis. For PD-L1 assay issues that affect combination patient selection, see the PD-L1 assays synthesis.
References
Each citation links to the original publication via DOI. The same records are searchable in the evidence library by title or DOI.
- Sen T, Rodriguez BL, Chen L, et al. Targeting DNA Damage Response Promotes Antitumor Immunity through STING-Mediated T-cell Activation in Small Cell Lung Cancer. Cancer Discov. 2019;9(5):646–661. doi:10.1158/2159-8290.CD-18-1020. ↩
- Domchek SM, Postel-Vinay S, Im SA, et al. Olaparib and durvalumab in patients with germline BRCA-mutated metastatic breast cancer (MEDIOLA). Lancet Oncol. 2020;21(9):1155–1164. doi:10.1016/S1470-2045(20)30324-7. ↩
- Schmid P, Im SA, Armstrong A, et al. BEGONIA: Phase 1b/2 study of durvalumab combinations in locally advanced/metastatic triple-negative breast cancer. Ann Oncol. 2023;34(suppl_2):S260–S261. ↩
Last reviewed: 2026-06-04. Researcher-layer synthesis page. Evidence grades follow the GRADE-adapted rubric defined at the top of this page.