161 Discussion and future perspectives patients (MEGALiT in Sweden (NCT04185831), IMPRESS in Norway (NCT04817956) and ProTarget in Denmark (NCT04341181)). Another example of international collaboration is the joint effort of DRUP and the Australian Molecular Screening and Therapeutics clinical trials & immunotherapy (MoST clinical trials). Both DRUP and MoST found in a considerable number of patients with genomic aberrations in the Cyclin D-CDK4/6 pathway treated with off-label CDK4/6 inhibitors no clinical benefit18. These results were reported as a pooled analysis of all cohorts across the two trials in which patients were treated with CDK4/6 inhibitors palbociclib and ribociclib, achieving a greater level of evidence than reporting of individual small cohorts. Within the DRUP trial, the vast majority of patients had a rare subtype of cancer. Either they had a rare cancer (an incidence less than 6 cases per 100,000 persons per year), or a common tumor type with a rare genomic aberration. Around 33% of enrolled patients in DRUP has a rare cancer, this group has the same overall clinical benefit rate (33%) of genomics-guided off label treatment as the group of patients with non-rare cancers19. Although it is considered a strength of DRUP that patients with rare cancers are offered an extra, potentially effective, treatment option, it also results in a large number of cohorts with just one or two patients enrolled. One of the challenges following from these small and “incomplete” cohorts is how to gain a sufficient level of evidence for these patients. DRUP has found several solutions for this issue, as described above. But even if cohorts are completed in stage 2 (24 patients), health authorities struggle to appreciate the evidence from small cohorts in a non-randomized non-controlled phase 2 trial. Especially the lack of a proper control group is difficult to overcome. Due to the rarity of these subgroups, conducting a randomized clinical trial is impossible. The use of historical controls is a generally accepted approach, but it has also proven to be nearly impossible to find correctly-matched controls, because patients qualifying for DRUP have exhausted standard-of-care treatments. Historical controls should ideally be matched by molecular subtype, however this information is often not reported in cohorts from the past. Finally, a commonly used endpoint is an intra-patient progression-free survival (PFS) ratio, defined as the PFS interval associated with molecularly guided therapy (PFS2) divided by the PFS interval associated with the last prior systemic therapy (PFS1), above 1.3 or, in some studies, above 1.33 or 1.520. Using this ratio, the patient serves as his/her own control. Among other issues leading to potential bias21, one difficulty in this approach is that the PFS1 data are retrospectively retrieved, while the PFS2 data are prospectively collected. Another ongoing challenge is to find the most accurate method of prioritizing different molecularly guided treatment options for individual patients. Growing experience with the results of tumor broad panel sequencing or WGS teaches us that tumor DNA often harbors more than one genomic aberration. A large pan-cancer analysis shows that tumors have a mean number of 5.7 candidate genomic driver events per patient8, likely occurring at different stages of tumor evolution. Some tumors may have multiple drivers occurring as early events in tumor development. But which genomic feature is the dominant driver that should be the target for treatment? It is possible that the answer lies in the administration of combinations of targeted anti-cancer agents. In the I-PREDICT trial, patients were treated with one or more 6
RkJQdWJsaXNoZXIy MTk4NDMw