Pathologic complete responses were improved when neoadjuvant atezolizumab was added to nab-paclitaxel followed by doxorubicin plus cyclophosphamide compared with placebo in combination with chemotherapy as treatment of patients with early triple-negative breast cancer without adding additional treatment burden to patients.
Pathologic complete responses (pCRs) were improved when neoadjuvant atezolizumab (Tecentriq) was added to nab-paclitaxel (Abraxane) followed by doxorubicin plus cyclophosphamide (AC) compared with placebo in combination with chemotherapy as treatment of patients with early triple-negative breast cancer (TNBC) without adding additional treatment burden to patients, according to findings from an analysis of patient-reported outcomes (PROs) of the IMpassion031 trial (NCT03197935) presented during the 2020 San Antonio Breast Cancer Symposium.1
Results showed that mean physical function at baseline, which was 91% in the atezolizumab cohort (95% CI, 89-93) and 90% in the placebo cohort (95% CI, 88-92), was high and remained comparable between both study arms. Both cohorts of patients experienced similar and clinically meaningful deterioration in physical function at the beginning of cycle 3 in the neoadjuvant period and that continued into cycle 5. However, both arms were able to rebound in the adjuvant period; starting in cycle 7, they began to achieve a gradual stability in terms of physical function.
Moreover, mean role function at baseline was high, at 89% in both the atezolizumab arm (95% CI, 86-93) and placebo arm (95% CI, 86-92), and this continued to remain comparable between both arms. Clinically meaningful deterioration in role function was observed in both cohorts during the neoadjuvant treatment period; both arms experienced the same magnitude of deterioration starting at the beginning of cycle 2 for those in the atezolizumab arm or cycle 3 in those in the placebo arm; this continued through cycle 5. Although patients continued to experience a negative impact in role function in the adjuvant setting, this began to stabilize in cycle 9 in the placebo/chemotherapy arm; this was observed while patients were not actively receiving treatment and were under observation.
“Patients on the atezolizumab arm of the trial did not completely recover or stabilize with respect to role function,” Elizabeth Ann Mittendorf, MD, PhD, Rob and Karen Hale Distinguished Chair in Surgical Oncology, Brigham and Women's Hospital; director of surgical research; director of the Breast Immuno-Oncology Program; and co-director of the Breast Cancer Clinical Research Program at Dana-Farber Cancer Institute, said during a presentation on the findings. “We attribute this to the fact that patients continue to receive atezolizumab, which required that they continued to come to the clinic for that therapy, thereby potentially impacting role function.”
Mean health-related quality of life (HRQoL) was also high at baseline in both the atezolizumab and placebo cohorts, at 79% (95% CI, 76-82) and 76% (95% CI, 73-79), respectively, with a clinically meaningful deterioration of comparable magnitude in both arms in the neoadjuvant period; this began during cycle 3 and continued through cycle 5. Both cohorts were able to rebound and achieve a gradual stability in HRQoL, beginning at cycle 6 in the adjuvant period. In the placebo arm, patients were able return to baseline mean values by cycle 7.
“Patients who received atezolizumab plus chemotherapy in the neoadjuvant treatment period on this study did not experience any additional detrimental impact on their day-to-day functioning or HRQoL compared with patients in the placebo arm in this study,” Mittendorf said. “Patients’ physical, role functioning, and HRQoL began to stabilize and return to baseline levels in both arms during the adjuvant treatment period.”
Data from the phase 3 IMpassion031 trial presented during the 2020 ESMO Virtual Congress showed that atezolizumab plus chemotherapy achieved a pCR of 57.6% compared with 41.1% in the placebo plus chemotherapy cohort (P = .0044); this benefit was observed regardless of PD-L1 expression status.2
Due to the largely asymptomatic nature of early breast cancer, investigators believe that TRAEs and their impact define a patient’s experience. Although toxicities associated with neoadjuvant chemotherapy are believed to be clinically manageable, they can still interfere with day-to-day life and lead to decreased adherence and even early treatment discontinuation.3 The IMpassion031 PRO assessment is the first to examine the treatment burden experienced by patients with early-stage breast cancer who have been received treatment with an immune checkpoint inhibitor.
“We used validated surveys that looked at TRAEs and [AE] bother, as well as surveys that assessed treatment impact, including role function, physical function, emotional, and social function, and HRQoL,” Mittendorf said.
The trial enrolled patients with TNBC who had a primary tumor of 2 cm or more. They were required to have cT2-cT4, cN0-cN3, and cM0 and they needed to have known PD-L1 status and an ECOG performance status of 0 or 1. Patients could not have received prior agents for the treatment or prevention of breast cancer.
A total of 333 patients were enrolled to the study and were randomized 1:1 to receive 840 mg of intravenous atezolizumab twice weekly plus nab-paclitaxel every 2 weeks followed by atezolizumab plus AC or placebo plus nab-paclitaxel followed by placebo plus AC. Following surgery, patients received either 11 doses of atezolizumab at 1200 mg every 3 weeks or were put on observation.
The co-primary end points of the study were pCR in both the intent-to-treat population and PD-L1–positive subpopulation. Secondary PRO end points included mean and mean changes from baseline in function (role and physical) and global health status/HRQoL.
Additional exploratory PRO end points consisted of mean and mean changes from baseline in terms of function (emotional, social, and cognitive), as well as disease and/or treatment-associated symptoms; the proportion of patients reporting each response at each assessment time point by treatment arm for the treatment of bother item served as another end point.
PRO completion rates for the 2 respective questionnaires, the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30; treatment-related adverse effects [TRAEs]), as well as the Functional Assessment of Cancer Therapy-General single item GP5 (FACT-G GP5; AE bother), remained high throughout treatment.
For EORTC QLQ-C30, upon treatment discontinuation, completion rates were 93% and 95% in the investigative and control arms, respectively, although they dropped to 67% and 66%, respectively, at 24 months. Similar completions rates were seen for FACT-G GP5, with 95% and 91% of each cohort, respectively, completing the questionnaire around the time of treatment discontinuation and dropping to 67% and 66%, respectively, at 24 months.
“This likely reflects the fact that at the time of this analysis, the median follow-up for the study was only 20 months,” noted Mittendorf.
With regard to TRAEs, such as fatigue, nausea, and vomiting, and diarrhea, a clinically meaningful worsening of AEs was noted in cycle 5 in both treatment arms, with the exception of pain, which was observed at cycle 4. This was found to stabilize in the adjuvant setting; the mean values in both cohorts at visits through cycle 16 were comparable to those reported at baseline for most of the symptoms assessed, with the exception of fatigue.
“It might suggest that there was not a complete stabilization in the atezolizumab [arm] with respect to fatigue, but I would highlight that the confidence intervals overlap, suggesting no differences between the arms of this study.”
In the neoadjuvant period, both arms reported similar levels of increased AE bother by visit. However, in the adjuvant period, no additional TRAE bother was noted in patients who were treated with atezolizumab/chemotherapy vs those who received placebo/chemotherapy who were under observation and not receiving treatment.