OBJECTIVES:  “Head-to-head” trial or direct comparison is the classical approach considered as gold standard to compare the efficacy, safety and additional overall benefit of 2 treatments. Indirect comparison may in some cases be the only option to compare interventions. Although, in theory, French health authority (HAS) accepts the indirect comparisons and a guideline for indirect comparison is published, it is not clear if they are accepted in practice. The aim of this study is to identify the number of indirect comparisons in oncology as well as their acceptability by HAS. METHODS:  HAS reports published between 01/01/2012 and 31/12/2016 for oncology products were extracted from HAS website. Only initial submission reports were included in this study. Generics and biosimilar products assessments were excluded. Then, indirect comparisons were identified in each report as well as the opinion of HAS on these comparisons when available. RESULTS:  292 reports for oncology products were extracted among which 67 were included in this study. Indirect comparisons were submitted only for 8 of the 67 products in addition to the head-to-head randomized clinical trials. HAS considered that indirect comparisons have a minor impact and they were not considered actually in the final assessment. Consistently, HAS questioned the value of these indirect comparisons because they were thought to have limitations due to period differences, potential heterogeneity of studies (population and patients’ management etc) as well as potential population selection that may be very different even though those heterogeneity criticisms were not robustly documented. CONCLUSIONS:  The use of indirect comparison is becoming unavoidable as it is almost impossible to generate comparative head-to-head data for all relevant interventions. Despite some products indirect comparisons availability for other HTA agencies, they are not filed to HAS. When filed they happen to have very little impact on the HAS assessment.