Background: In the neoadjuvant setting, blockade of HER2 plus use of an aromatase inhibitor in patients with HER2-positive and oestrogen receptor (ER)-positive breast cancer leads to a pathological complete response in 21% of patients. Convergence of HER2 and ER signals on RB1 suggests that a combined pharmacological intervention directed to these targets could be synergistic. To test this approach, we combined palbociclib to block RB1, fulvestrant to block ER, and trastuzumab with pertuzumab to block HER2 in patients with HER2-positive, ER-positive breast cancer. Methods: NA-PHER2 is a multicohort, open-label, exploratory, phase 2 study done at seven sites in Italy. Patients were eligible for the first cohort if they had previously untreated, histologically confirmed, unilateral, invasive, HER2-positive, ER-positive breast cancer and were suitable for neoadjuvant therapy. Patients were treated every 3 weeks with intravenous trastuzumab (8 mg/kg loading dose followed by 6 mg/kg) and intravenous pertuzumab (840 mg loading dose in the first cycle and then at 420 mg) for six cycles plus oral palbociclib (125 mg once a day for 21 days in a 4-week cycle) and intramuscular fulvestrant (500 mg) every 4 weeks for five cycles. The coprimary endpoints were change from baseline in Ki67 expression at 2 weeks of treatment and at surgery (16 weeks after treatment) and changes in apoptosis from baseline to surgery. Secondary endpoints were clinical objective response (according to modified Response Evaluation Criteria in Solid Tumors) and pathological complete response. All patients who met eligibility criteria were assessed for the primary and secondary endpoints. All patients who received at least one cycle of therapy were assessed for safety. This trial is registered with ClinicalTrials.gov, number NCT02530424. The trial is ongoing and two further cohorts are being enrolled. Findings: Between May 20, 2015, and Feb 8, 2016, we enrolled 36 patients, of whom one was deemed ineligible for the study and five were found to be HER2-negative on retrospective analysis. Thus, 35 patients were included in safety analyses and 30 were assessed for the primar y and secondary endpoints. At baseline, geometric mean Ki67 expression was 31·9 (SD 15·7), versus 4·3 (15·0) at week 2 (n=25; p <0·0001) and 12·1 (20·0) at time of surgery (n=22; p=0·013). The geometric mean for apoptosis was 1·2 (SD 0·3) at baseline versus 0·4 (0·4; p=0·019) at surgery. A clinical objective response immediately before surgery was achieved by 29 (97%; 95% CI 83-100) of 30 patients. At surgery, eight (27%; 95% CI 12-46) patients had a pathological complete response in breast and axillary nodes. The most frequent grade 3 adverse events were neutropenia (ten [29%]), diarrhoea (five [14%] ), and stomatitis, increased alanine aminotransferase, and hypersensitivity reactions (one [3%] of each event). No grade 4 or serious adverse events were recorded in the study and there were no deaths. Interpretation: The combination of palbociclib, fulvestrant, trastuzumab, and pertuzumab had a significant effect on the expression of Ki67 at 2 weeks and at surgery. Triple targeting of ER, HER2, and RB1 in HER2-positive and ER-positive breast cancer could be an effective chemotherapy-free treatment strategy. Further clinical testing and additional molecular characterisation is necessary, not only in hormone receptor-positive tumours but also in tumours without HER2 amplification. Funding: Pfizer and Roche. © 2018 Elsevier Ltd.