The neoadjuvant combination of palbociclib (Ibrance), pertuzumab (Perjeta), fulvestrant (Faslodex), and trastuzumab (Herceptin), cut expression of Ki67 and induced an overall clinical response of 29% in women with ER-positive/HER2-positive breast cancer.
Luca Gianni, MD
The neoadjuvant combination of palbociclib (Ibrance), pertuzumab (Perjeta), fulvestrant (Faslodex), and trastuzumab (Herceptin), cut expression of Ki67 and induced an overall clinical response of 29% in women with ER-positive/HER2-positive breast cancer.1
Palbociclib was used to block RB1, fulvestrant to block ER, and trastuzumab with pertuzumab to block HER2.
The NA-PHER2 study is a multicohort, open-label, exploratory, phase II study conducted at 7 sites in Italy. Investigators recruited 36 patients with previously untreated, histologically confirmed, unilateral, invasive, ER-positive/HER2-positive cancer. Eligible patients were suitable for neoadjuvant therapy.
Patients were treated every 3 weeks with a loading dose of 8 mg/kg of intravenous trastuzumab followed by a 6 mg/kg dose, an 840-mg loading dose of intravenous pertuzumab in the first cycle then a 420-mg dose for 6 cycles, plus 125 mg of oral palbociclib once daily for 21 days in a 4-week cycle, and 500 mg of intramuscular fulvestrant every 4 weeks for 5 cycles. The treatment period lasted from May 2015 to February 2016.
The coprimary endpoints were change from baseline in Ki67 expression after 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 and pathologic complete response (pCR).
In this first cohort, 30 patients were included in endpoint analysis and 35 were included in the safety analysis. Investigators are conducting 2 additional cohorts and an ongoing ancillary study of tissue samples collected before and during treatment to learn more about this treatment regimen.
At baseline, geometric mean Ki67 expression was 31.9 (standard deviation (SD), 15.7) compared with 4.3 (SD, 15.0) at week 2 (n = 25; P <.0001) and 12.1 (SD, 20.0) at time of surgery (n = 22; P = .013). The geometric mean for apoptosis was 1.2 (SD, 0.3) at baseline versus 0.4 (P = .019) at surgery.
Immediately prior to surgery, investigators observed a clinical objective response immediately in 29 of 30 patients (97%; 95% CI, 83-100). At surgery, 8 patients (27%; 95% CI, 12-46) had a pCR in breast and axillary nodes. Twenty-two patients had residual invasive disease at surgery.
Ki67 could not be assessed due of a lack of tumor cells in 5 patients collected at week 2. Three of those patients later achieved pCR, which, according to the investigators, suggests that they experienced an almost immediate tumor breakdown.
Investigators acknowledged that they cannot “set a new standard for therapy” because the study was exploratory, did not have a control arm, and the patient population was small.
“However, our findings support further investigation of this chemotherapy-free regimen in randomized clinical trials and in larger cohorts of women with ER-positive, HER2-positive breast cancer at different stages of disease, including in the metastatic setting,” wrote lead author Luca Gianni, MD, director of Medical Oncology at the San Raffaele Hospital in Milan, and colleagues. “A triple block of HER2, ER, and RB1 is promising in HER2-positive and ER-positive disease and could work also in the absence of HER2 amplification or overexpression in ER-positive tumors.”
There were no incidents of grade 4 or serious adverse events (AEs) recorded in the study. Diarrhea (74%) and neutropenia (69%) were the most common any-grade AEs. The most frequent grade 3 AEs were neutropenia (29%) and diarrhea (14%).
There were no deaths in the study.
Writing in an accompanying editorial, Xavier Pivot, MD, PhD, and David G. Cox, PhD, wrote that using biological indicators like Ki67 to assess efficacy could be an interesting way to identify the activity of treatment early in the neoadjuvant setting.2 However, the relationship between biological modification and treatment efficacy has not been established.
“The clinical value of such indicators is debatable and the effort needed to include biomarkers in future study designs is most likely to be considerable,” they wrote.
Nonetheless, they concluded that the “encouraging” activity and safety associated with this regimen is worth exploring further.
“The NA-PHER2 study reflects what might be the future: optimum study design combining extensive biological assessment, limited population accrual, early exposure in a neoadjuvant setting, and original endpoints based on tumor biological changes as indicators of early activity,” wrote Pivot and Cox. “On the other hand, such modern designs need prudence with respect to the effect of the clinical trial on cure status, and vigilance addressing access to standard well-established strategies for cure.”