作者
Jindřich Fínek,Regina Demlová,Kateřina Kopečková,Tomáš Büchler,Bohuslav Melichar,Jana Prausová,Alexandr Poprach,Tereza Hrnčiarová,T. Mlcoch,Tomáš Doležal
摘要
Background: Efficacy data of treatment sequences in mRCC are rare. In the Czech Republic, data on efficacy and safety of all targeted therapies for mRCC are collected in the RENIS patient registry. Thus, RENIS provides data from the real-world clinical practice. The aim of this study was to compare outcomes of selected treatment sequences in mRCC while adjusting for differences in patient characteristics using inverse propensity score weighting method (IPWS). Methods: Data of mRCC patients treated using most common treatment sequences were collected in RENIS between 06/2007 and 02/2018. Overall survival (OS) and progression free survival (PFS) were evaluated. Baseline characteristics were balanced using IPWS. The propensity score was evaluated with nominal logistic model to balance Eastern Cooperative Oncology Group (ECOG) performance status, time from diagnosis to first treatment, nephrectomy, Memorial Sloan-Kettering Cancer Center (MSKCC) score, and age. Median and confidence intervals (CI) were derived from IPWS weighted Kaplan-Meier curves that were compared using log-rank test. Results: Overall, 745 patients in five treatment sequences were included and analysed. Differences in OS were significant (p < 0.001) with sunitinib→axitinib→everolimus and sunitinib→axitinib sequences associated with improved survival over other sequences. These differences could be also linked to time when these drugs were introduced. PFS did not differ between sunitinib and pazopanib used as first line treatments (p = 0.44) but the PFS in the second line differed significantly (p = 0.035) (Table). Table: Results of OS and PFS for examined sequences computed using inverse propensity score weighting. Sunitinib → Axitinib → Everolimus sequence should be assessed with caution due to immortal time bias.Table: 892PParameterTreatment sequencenMedian95% CI for medianpLower limitUpper limitOS (months)Sunitinib → Everolimus31226.323.829.0<0.001Sunitinib → Axitinib15447.631.350.5Sunitinib → Axitinib → Everolimus†11446.044.750.9Pazopanib → Everolimus7432.930.933.8Pazopanib → Sunitinib9127.226.728.8PFS 1st line (months)Sunitinib58010.29.810.80.440Pazopanib1659.28.510.0PFS 2nd line (months)Sunitinib → Everolimus3125.85.16.30.035Sunitinib → Axitinib2686.46.07.1Pazopanib → Everolimus745.14.15.5Pazopanib → Sunitinib915.35.17.1 Open table in a new tab Conclusions: Improved outcomes were associated with sequences using second-line axitinib over those using second-line sunitinib or everolimus in a cohort of patients from a national registry. Legal entity responsible for the study: Jindrich Finek. Funding: Value Outcomes. Disclosure: J. Finek: Honoraria: Amgen, BMS, Roche, Bayer, Teva, MSD, Merck, Sanofi, Pierre Fabre. R. Demlova: Honoraria for lectures: Bayer. K. Kopeckova: Honoraria for lectures: Novartis; Travel grants: Pfizer, Bayer. T. Buchler: Honoraria for advisory boards and/or lectures: Novartis, Pfizer, Bayer, Roche. B. Melichar: Honoraria for advisory boards and/or lectures: Novartis, Pfizer, Glaxo Smith Kline, Roche, Bayer. A. Poprach: Honoraria for lectures: Novartis, Bristol-Myers Squibb, Roche, Bayer. T. Hrnciarova, T. Mlcoch, T. Dolezal: Employee of Value Outcomes. All other authors have declared no conflicts of interest.