Subgroup analysis
Considering the high heterogeneity shown in the pooled results, subgroup analysis was performed to identify the sources of the heterogeneity. We chose several common factors that related to effectiveness including age, ECOG performance status (PS), primary site of tumor, region, and HER2 expression status as classification variables. However, only ORR data were used for subgroup analysis due to the limited data.
The results showed (Figure 4 A-D) that none of the differences in age, ECOG PS, primary tumor type or region was a source of heterogeneity, and there was no significant difference between the subgroups (p=0.941; p=0.748; p=0.625, p=0.734, respectively). The HER2 status subgroup analysis showed (Figure 4E) that the ORR of HER2-positive and HER2-low patients were significant different (39% vs 19%, p=0.037). The heterogeneity of ORRs in both the HER2-postive group (I2=35.5%, p =0.171) and HER2-low group (I2=0.0%, p =0.843) was reduced significantly compared to the pooled results (I2=53.0%,p =0.037), indicating that HER2 expression status was one of the factors leading to heterogeneity.
Given T-DXd was the only drug which has been used as intervention regimen in more than one study in this meta-analysis, we further calculated the efficacy of T-DXd in patients with HER-2 overexpression. As the results showed (Figure 4 F-I), the pooled ORR and DCR were respectively 43% (95% CI: 36%-49%) and 83% (95% CI: 79%-88%). The pooled mOS and mPFS were respectively 12.36 months (95% CI: 10.53-14.19 months) and 5.60 months (95% CI: 4.59-6.61 months). And no heterogeneity was observed among these results.