For patients (pts) with advanced HR+/HER2- breast cancer various treatment options exist. Endocrine therapy or chemotherapy are the recommended 1st-line treatment options according to international guidelines. With comparable efficacy, it is of utmost importance to identify the treatment that has the least negative impact on the pts' quality of life (QoL). Randomized studies to determine the pts' preference (pref) for equi-effective treatment concepts, are lacking. IMPROVE compared both therapeutic concepts by assessing the pts' pref for either a combined antihormonal approach (everolimus /exemestane (E/E)) or chemotherapy (capecitabine /bevacizumab (C/B)).

In total, 77 pts were recruited from 10/14 until 04/17 at 26 sites in GER. Upon diagnosis of locally advanced, inoperable or metastatic disease, pts were randomized to receive 1st-line C/B until disease progression followed by 2nd-line E/E (Arm A) or vice versa (Arm B). Primary objective was the patient-reported pref for either treatment protocol 12 weeks after switching therapy. Key secondary endpoints include PFS, OS, safety and QoL. Descriptive statistics were used to to analyse data, PFS and OS were calculated by using the Kaplan-Meier method.

Baseline characteristics were well balanced with a slightly shorter disease-free interval after primary diagnosis and more prior treatments in Arm B.In Arm A [95% CI], 39% [13.9-68.4] vs 23% [5.0-53.8] of pts preferred E/E compared to C/B, 23% [5.0-53.8] were undecided. In Arm B, 56% [30.8-78.5] vs 11% [1.4-34.7] preferred C/B, 22% [6.4-47.6] undecided. Overall, 42% [24.5-60.9] vs 23% [9.6-41.1] of pts preferred C/B, 23% [9.6-41.1] were undecided.

Physicians' pref had a tendency for C/B treatment (Arm A, 62% [31.6-86.1] vs 39% [13.9-68.4] for C/B. Arm B, 56% [30.8-78.5] vs 28% [9.7-53.5] for C/B, no pref 17% [3.6-41.4]).

Median 1st-line PFS [months, 95% CI] was 11.1 [7.8-18.0] for C/B (Arm A) vs 3.5 [2.7-5.5] for E/E (Arm B). Median 2nd-line PFS was 3.7 [2.4-7.8] for E/E (Arm A) vs 3.6 [2.3-5.5] for C/B (Arm B). Median OS [months, 95% CI] was 28.8 [19.7-NA] (Arm A) and 24.7 [13.9-28.8] (Arm B). 73.0% and 52.6% (1st and 2nd line, C/B) vs 54.1% and 52.9% (1st and 2nd line, E/E) of pts developed grade 3/4 AEs. Most common grade 3/4 AEs (%) were hand-foot syndrome (18.9), fatigue (17.6), hypertension (13.5) for C/B and anaemia (23.5), fatigue, dyspnoea, cough (each 17.6) for E/E (either line). No treatment-related deaths occurred. Patient-reported QoL (EORTC-QLQC30) and treatment satisfaction were not significantly different between arms in either treatment phase.

In the IMPROVE study, pts had no pref for either endocrine therapy or immunochemotherapy. Overall, there was a tendency in favour of the chemotherapeutic approach (C/B), which was in line with the therapy pref reported by the physicians. C/B was found to have slightly better efficacy results but at the cost of a higher frequency of grade 3/4 AEs, bearing in mind the difference in duration of therapy between the two regimens. Patient-reported QoL, however, was similar in both arms.

Citation Format: Kurbacher CM, Söling U, Hahn A, Chiabudini M, Maintz C, Rieger L, Falkenstein J, Runkel E, Potthoff K, Decker T. Final results from IMPROVE: A randomized, controlled, open-label, cross-over phase IV study to determine the patients' preference for either combined endocrine therapy (exemestane plus everolimus) or immunochemotherapy (capecitabine plus bevacizumab) as first line treatment for advanced HR+/HER2- breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-21-11.