Table 2. Definitions of the treatment categories.

Category		Definition
According to approval status
Approved for cancer type analysed	Treatment approved by the FDA for the tumour type being analysed
Approved for other cancer type	Treatment approved by the FDA for other tumour but not for cancer type being analysed
Under development	Treatments in development, tested in phase II, II and IV clinical trials recruiting patients during 2016. These clinical trials (specifying altered genes or pathways within the inclusion criteria) were identified using ClinicalTrials.gov (https://clinicaltrials.gov/).

According to clinical benefit
Potential clinical benefit (PCB)	Treatments with evidence for a potential clinical benefit (either associated with a FDA recognized biomarker, present in guidelines or associated with strong clinical evidence). e.g. erlotinib for EGFR-activating mutations in NSCLC, Larotrectinib for NTRK1-fusion solid tumours
Lack of potential clinical benefit (lack of PCB)	Treatments associated with a resistance (e.g. Erlotinib in EGFR T790M mutant NSCLC) or standard of care treatments with no evidence for a potential clinical benefit due to absence of an alteration in the patient (e.g. Pembrolizumab in MSS CRC or in PD-L1 negative NSCLC)
Unknown clinical benefit (unknown PCB)	Treatments with no strong evidence of efficacy (i.e. based only on preclinical data) or with contradictory evidence either found in the literature or based on patient pathway analysis (i.e. contradictory clinical benefit information found for the treatment after comparing the therapeutic impact of alterations detected by next-generation sequencing to those detected in “Package Plus” (IHC) in a sample)
Without treatment	No alternative treatment can be recommended (i.e. no molecular alteration that could predict response or resistance to treatment was detected in the sample)

Categorization according to approval status takes into consideration if the treatment has been FDA approved and in which indications, or if it is still in clinical trial investigation.

Categorization according to clinical benefit evaluates the strength of evidence and the type of response to a treatment based on the presence of a specific molecular alteration.

Note that these 2 categories are independent. A treatment being classified according to the approval status does not automatically qualify it to any of the categories in clinical benefit. Indeed, a FDA approved drug can be classified as “unknown” clinical benefit if there is no strong evidence of clinical benefit to the aberration detected in the sample, or if there is contradictory evidence based on patient pathway analysis.

MSS: microsatellite stable.