Skip to main content
. 2021 Feb 12;7:11. doi: 10.1038/s41523-021-00222-y

Fig. 7. Clinical outcomes for non-VM and VnLM versus VLM in the second-line setting.

Fig. 7

a Forest plots of PFS, OS, DoCB and CBR. *Fixed effect for trial was fitted in all models. Fixed-effect model was fitted to the AI SERD 250 mg and SERD 500 mg data; random effects for trial were included in the models for all data. §“one−stage” fixed−effects logistic regression models. b Projected probability of PFS, OS and DoCB. Median (95% CI) PFS/OS/DoCB in months. AI aromatase inhibitor, CBR clinical benefit rate, DoCB duration of clinical benefit, HR hazard ratio, n number of patients, non-VM non-visceral metastases, PFS progression-free survival, OR odds ratio, OS overall survival, SERD selective estrogen receptor degrader, SERM selective estrogen receptor modulator, VLM visceral liver metastases, Non-VM non-visceral metastases, VnLM visceral non-liver metastases. VLM vs VnLM vs non-VM. PFS: AI: Interaction test p = 0.15; SERD (250): Interaction test p = 0.18; SERD (500): Interaction test p = 0.05; All: Interaction test p = 0.06. OS: AI: Interaction test p = 0.66; SERD (250): Interaction test p = 0.98; SERD (500): Interaction test p = 0.98; All: Interaction test p = 0.78. DoCB: AI: Interaction test p = 0.59; SERD (250): Interaction test p = 0.10; SERD (500): Interaction test p = 0.11; All: Interaction test p = 0.03.CBR: AI: Interaction test p = 0.90; SERD (250): Interaction test p = 0.20; SERD (500): Interaction test p = 0.33; All: Interaction test p = 0.47.