Background: HoR+ breast cancer (BC) is an heterogeneous disease, comprising 4 molecular subtypes (i.e.Luminal A [LumA], Luminal B [LumB]), HER2-Enriched [HER2E] and Basal-like [BL]). In HoR+ MBC, non-LumA subtypes appear to be associated with poorer survival outcome compared to LumA. Here, we present a meta-analysis to validate the prognostic value of IS inHoR+ MBC. Methods: We performed a systematic review and trial-level meta-analysis of all available prospective phase II/III trials in HoR+ MBC where the prognostic role of the IS was assessed in terms of progression-free survival (PFS) or time-to-progression (TTP), if the former was not available. Hazard ratios (HR) and 95% confidence intervals (CI) had to be available or computable. Subgroup analyses according to treatment, menopausal and HER2 status were also performed. The random-effect model of Der Simonian and Laird was applied. Heterogeneity was assessed through Cochran’s Q and I2. Funnel plot with Egger’s and Begg’s test for publication bias, multiple sensitivity analyses and risk of bias analysis were also conducted. Revman 5.4 and R 3.6.1 for MacOS X were used for statistical analyses. Tests were two-sided and the threshold for significance was set at p<0.05. Results: Overall, 360 records were screened from Pubmed and the Cochrane CENTRAL. Six randomized phase III trials of either endocrine therapy (ET)±target therapy (TT) or chemotherapy (only 1 study) and 1 single arm phase II trial of ET+TT were included, for a total of 2,536 patients (pts). 1,401 (55.2%) pts presented with non-LumA BC and 1,135(44.8%) presented with LumA BC. Within non-LumA BC, 574 (50.0%) were LumB, 280 (20.0%) HER2E and 62 (4.4%) BL. Compared to LumA, non-LumA tumors were associated with worse PFS/TTP (HR: 1.77, 95% CI: 1.54 - 2.05, p<0.001), independently of HER2 (HER2+ vs. HER2-negative, psubgroup difference[psub]=0.16), systemic treatment (ET vs. ET+TT vs. CT, psub=0.96) and menopausal status (pre vs. post, psub=0.12). Compared to LumA, each IS showed a statistically significant poorer outcome (psub=0.01), with LumB (HR: 1.46, 1.21 -. 1.77) showing a better prognostic effect than HER2E (HR: 2.39, 1.78 - 3.21) and BL (HR:2.67, 1.61 - 4.43). Heterogeneity was moderate (I2=61%, pheterogeneity[pH]<0.001). Sensitivity analyses confirmed the robustness of the main result, which did not significantly vary by removing the most influential cases sequentially or concomitantly. No publication bias was observed (Egger’s p=0.492; Begg’s p=0.719) and risk of bias analysis did not raise concerns regarding studies’ internal validity. Conclusions: Compared to LumA, the other subtypes are consistently associated with poorer survival outcome in HoR+ MBC, independently of HER2, treatment and menopausal status. Future trial designs in HoR+ MBC should consider this biological heterogeneity.

Table 1.

Included studies characteristics

Study ReferenceTrial NameBreast CancerSubgroupMenopausal StatusTrial TypeTreatmentNon-Luminal ALuminal A
Prat A. et al. The Oncologist 2019 BOLERO2 HR+/HER2-neg Postmenopausal Randomized Everolimus+exemestane vs exemestane 139 122 
Jørgensen C.L.T. et al. Acta Oncologica 2014 DBCG Mainly HR+/HER2-neg Mixed Randomized Docetaxel+gemcitabine vs docetaxel 186 84 
Prat A. et al. JAMA Oncol 2016 EGF30008 HR+/HER2+ and neg Postmenopausal Randomized Letrozole+lapatinib vs letrozole 424 377 
Prat A. et al. J Clin Oncol 2021 MONALEESA 2 HR+/HER2-neg Postmenopausal Randomized Ribociclib+letrozole vs letrozole 
Prat A. et al. J Clin Oncol 2021 MONALEESA 3 HR+/HER2-neg Postmenopausal Randomized Ribociclib+fulvestrant vs fulvestrant 618 542 
Prat A. et al. J Clin Oncol 2021 MONALEESA 7 HR+/HER2-neg Premenopausal Randomized Ribociclib+AI/TAM vs AI/TAM 
Ciruelos E. et al. Clin Can Res 2020 PATRICIA HR+/HER2+ Postmenopausal Non-Randomized Palbociclib+/-letrozole+trastuzumab 34 10 
        
Study ReferenceTrial NameBreast CancerSubgroupMenopausal StatusTrial TypeTreatmentNon-Luminal ALuminal A
Prat A. et al. The Oncologist 2019 BOLERO2 HR+/HER2-neg Postmenopausal Randomized Everolimus+exemestane vs exemestane 139 122 
Jørgensen C.L.T. et al. Acta Oncologica 2014 DBCG Mainly HR+/HER2-neg Mixed Randomized Docetaxel+gemcitabine vs docetaxel 186 84 
Prat A. et al. JAMA Oncol 2016 EGF30008 HR+/HER2+ and neg Postmenopausal Randomized Letrozole+lapatinib vs letrozole 424 377 
Prat A. et al. J Clin Oncol 2021 MONALEESA 2 HR+/HER2-neg Postmenopausal Randomized Ribociclib+letrozole vs letrozole 
Prat A. et al. J Clin Oncol 2021 MONALEESA 3 HR+/HER2-neg Postmenopausal Randomized Ribociclib+fulvestrant vs fulvestrant 618 542 
Prat A. et al. J Clin Oncol 2021 MONALEESA 7 HR+/HER2-neg Premenopausal Randomized Ribociclib+AI/TAM vs AI/TAM 
Ciruelos E. et al. Clin Can Res 2020 PATRICIA HR+/HER2+ Postmenopausal Non-Randomized Palbociclib+/-letrozole+trastuzumab 34 10 
        

Citation Format: Francesco Schettini, Olga Martínez-Sáez, Nuria Chic, Fara Brasó-Maristany, Patricia Galván, Debora Martínez, Laia Paré, Maria Vidal, Barbara Adamo, Montserrat Muñoz, Tomás Pascual, Eva Ciruelos, Charles M Perou, Lisa A Carey, Aleix Prat. Prognostic value of intrinsic subtypes (IS) in hormone receptor-positive (HoR+) metastatic breast cancer (MBC): A systematic review and meta-analysis of prospective trials [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P4-07-08.