Real-world outcomes of doxorubicin and paclitaxel in advanced Kaposi sarcoma: A TriNetX retrospective cohort study.

11577 Background: For decades, pegylated liposomal doxorubicin (PLD) and paclitaxel have been standard antineoplastic therapies for advanced Kaposi sarcoma (KS). However, contemporary estimates of their efficacies remain limited. This limit is particularly profound in HIV-associated KS, where advances in antiretroviral therapy and supportive care reduce the relevance of historical outcomes. Methods: In January 2026, the TriNetX database (171 global healthcare organizations) was queried to identify patients with KS (ICD-10-CM code C46 and ICD-O-3 code 9140/3) who received doxorubicin or paclitaxel since 2010. Patients with a history of lymphoma, cervical cancer, anal cancer, or solid organ transplantation were excluded. We extracted demographics, overall survival (OS), and time to next-line treatment or death (TTNT-D) to obtain estimated progression-free survival (ePFS). Outcomes were measured from index treatment initiation (first recorded receipt of doxorubicin or paclitaxel after KS diagnosis). Lastly, HIV-based subgroup analyses were conducted with 1:1 propensity matching by age, sex, race, and prior antineoplastic treatments. Kaplan-Meier analysis was done with significance set at p<0.05. Results: We analyzed 935 KS patients treated with doxorubicin (including PLD) and 379 treated with paclitaxel. The patients treated with doxorubicin (median age: 45 years, 81.0% male, 48.3% HIV+) had 2- and 5-year OS of 77.5% and 70.0% and ePFS of 63.0% and 56.36% respectively. The patients treated with paclitaxel (median age: 50 years, 78.5% male, 42.7% HIV+) had 2- and 5-year OS of 71.3% and 62.4% and ePFS of 53.8% and 45.1% respectively. Within the doxorubicin cohort, HIV+ patients were younger (median age 40 vs 56 years), more often male (95.0% vs 64.7%), and had higher proportions of African American and Hispanic patients (36.4% and 14.7% vs 12.3% and 7.1%). A matched analysis of HIV+ vs. HIV- patients showed 2-year OS of 86.2% vs 71.7% (p=0.003) and 2-year ePFS of 74.85% vs 52.19% (p<0.0001). Within the paclitaxel cohort, HIV+ patients were also younger (median age 40 vs 64 years), more often male (96.5% vs 63.8%), and had higher proportions of African American and Hispanic patients (39.3% and 15.0% vs 9.0% and 5.0%). A matched analysis of HIV+ vs. HIV- patients showed 2-year OS of 60.3% vs 74.1% (p=0.151) and 2-year ePFS of 38.0% vs 57.7% (p=0.023). A matched analysis of doxorubicin vs. paclitaxel in HIV+ patients showed 2-year OS of 86.9% vs 73.0% (p=0.007) and ePFS of 77.7% vs 67.9% (p=0.069). Conclusions: This large real-world analysis provides a contemporary assessment of doxorubicin and paclitaxel in advanced KS. HIV+ status was associated with higher OS and ePFS in the doxorubicin cohort, whereas HIV- status was associated with higher ePFS in the paclitaxel cohort. Limitations include the retrospective design and imperfect reliability of TTNT-D as a surrogate for PFS.