Why This Study Matters
The source paper states that the standard of care for Stage II–III HER2-positive breast cancer is neoadjuvant multi-agent chemotherapy plus trastuzumab and pertuzumab (HP), yielding pCR rates of 50–60% and 3-year disease-free survival of approximately 92%, but that multi-agent chemotherapy is associated with substantial toxicity. EA1181 tests de-escalation to a single-agent taxane plus HP (THP) and is described by the authors as the first preoperative trial in HER2-positive breast cancer powered to evaluate survival outcomes after a de-escalated regimen. This report covers key secondary objectives: pCR rates and factors associated with pCR.
CompassHER2 pCR (EA1181) is the largest prospective study of neoadjuvant THP in Stage II- IIIa HER2-positive breast cancer.
— Tung et al., J Clin Oncol, Discussion
Study Design
EA1181 is a prospective, single-arm, NCI-sponsored trial coordinated by ECOG-ACRIN with participation from Alliance, SWOG and NRG. Patients received 12 weeks of HP with one of three taxanes: weekly paclitaxel × 12 (80 mg/m²), docetaxel every 3 weeks × 4 (75 mg/m²), or weekly nab-paclitaxel × 12 (125 mg/m²). Surgery occurred within 18 weeks of taxane initiation. Patients achieving pCR (ypT0/is ypN0) completed one year of HP without further adjuvant chemotherapy. The primary endpoint — 3-year recurrence-free survival among patients achieving pCR — will be reported when data mature. A HER2DX pCR score (Reveal Genomics) was determined centrally on diagnostic biopsies from 569 participants.
2,175 patients enrolled (March 2020–October 2023); 2,141 received at least one dose of THP and were evaluable; 781 ER-negative and 1,394 ER-positive enrolled; 569 tumors with a HER2DX pCR score; single-arm, no randomization; number of sites not specified in source.
Patient Population
Eligible patients had HER2-positive operable Stage II or IIIa breast cancer (T4 and N3 excluded); T2N0 tumors were capped at 20% of accrual. Median age was 55 years (range 22–88). 46% were node-positive and 89% had cT2 or cT3 tumors; 56% were grade 3. Paclitaxel was the initial taxane in about two-thirds of patients, docetaxel in about one-third, and nab-paclitaxel in 2%. About 7.6% of patients switched taxane and 17% had a dose reduction; 88% completed 4 cycles of THP, and 2,028 (95%) underwent surgery per protocol.
Pathologic Complete Response (Key Secondary Objective)
Overall pCR was 43.8% (95% CI, 41.6%–45.9%; n = 2,141). By ER status:
| Group | pCR rate | 95% CI |
|---|---|---|
| All evaluable (n = 2,141) | 43.8% | 41.6%–45.9% |
| HER2+/ER-negative (n = 774) | 63.7% | 60.2%–67.1% |
| HER2+/ER-positive | 32.5% | 30.0%–35.0% |
| ER-positive, ER 1–10% (n = 136) | 62.5% | 53.8%–70.6% |
| ER-positive, ER 11–70% (n = 281) | 51.6% | 45.6%–57.6% |
| ER-positive, ER >70% (n = 950) | 22.5% | 19.9%–25.3% |
The source text reports the ER-positive pCR as 32.5%; the abstract and some secondary coverage give 32.4%, and source Table 3 reports 32% (30–35%).
Subgroup Analyses: Predictors of pCR (Multivariable)
| Factor (reference) | All (n = 2,141) | HER2+/ER− | HER2+/ER+ |
|---|---|---|---|
| HER2 IHC 3+ (vs 2+/ISH+) | 6.14 (4.30–8.75) | 6.31 (3.63–10.94) | 6.06 (3.79–9.69) |
| Paclitaxel (vs docetaxel) | 1.50 (1.22–1.84) | 1.86 (1.34–2.58) | 1.30 (0.99–1.71) |
| T3 (vs T1) | 0.64 (0.42–0.96) | 0.47 (0.25–0.90) | 0.76 (0.44–1.31) |
| ECOG PS 1 (vs 0) | 0.78 (0.58–1.07) | 0.54 (0.33–0.88) | 0.99 (0.67–1.47) |
| ER 1–10% (vs >70%) | 3.09 (2.02–4.71) | — | 2.99 (1.95–4.60) |
| ER 11–70% (vs >70%) | 2.75 (2.03–3.72) | — | 2.68 (1.97–3.64) |
| PR 0% (vs >10%) | 2.22 (1.68–2.94) | — | 2.41 (1.77–3.29) |
| PR 1–10% (vs >10%) | 1.52 (1.08–2.14) | — | 1.58 (1.09–2.29) |
Values are odds ratios for pCR (95% CI), from source Table 2. Among ER-negative patients with cT3 tumors (15% of the cohort), pCR was lower but still 59%, per the authors.
HER2DX pCR Score (Analytic Cohort, n = 569)
| HER2DX group | All: n | All: pCR (95% CI) | ER−: pCR (95% CI) | ER+: pCR (95% CI) |
|---|---|---|---|---|
| High | 183 | 68% (60–74%) | 70% (62–77%) | 58% (41–74%) |
| Medium | 159 | 67% (59–74%) | 74% (62–84%) | 61% (50–71%) |
| Low | 227 | 19% (14–25%) | 31% (9–61%) | 18% (13–24%) |
| P-value | <0.001 | 0.010 | <0.001 |
In multivariable analysis including clinicopathologic factors, high (OR 2.75, 95% CI 1.36–5.58) and medium (OR 4.17, 95% CI 2.35–7.40) HER2DX scores were independently associated with pCR versus low. The authors note pCR rates were similar between high and medium groups, and that T stage and PR status lost significance once the score was added, possibly reflecting reduced power in the smaller subset.
Safety Profile
Adverse-event rates are not specified in source; this secondary analysis reports pCR and its predictors. The source notes that fewer than 1% of patients progressed during neoadjuvant THP and that 5% withdrew from study during neoadjuvant treatment or before surgery.
Interpretation and Broader Context
The authors state the observed pCR rates were virtually identical to those anticipated in the trial design and to previous smaller trials of single-agent taxane with HP, with the exception of the smaller Daphne feasibility study (98 patients; pCR 56.7%). They suggest the HER2DX score may be most useful for identifying low-score patients (two-thirds of ER-positive patients in this subset) who may need longer or additional therapy, while noting that the benefit of escalation remains unclear. These findings do not establish that de-escalated therapy is equivalent to multi-agent chemotherapy.
Limitations
Limitations noted in the source: EA1181 is a single-arm trial, so there is no randomized comparator and the pCR rates cannot show that THP is equivalent to multi-agent chemotherapy. Survival outcomes, which will determine whether de-escalation is adequate and include the primary endpoint of 3-year recurrence-free survival, are not yet available. The HER2DX analysis covers only 569 tumors, and the authors note that the smaller subset may have had reduced power to detect effects of T stage and PR status once the score was added.