Collectively, rare cancers (RC) account for ~25-30% of all cancer diagnoses and 25% of cancer deaths, are associated with poor outcomes and less evidence-based treatment. We designed the WEHI-Stafford Fox Rare Cancer Program (SFRCP) to allow the analysis of data and tissue of RC. We have laboratory processes for processing of samples, including the generation of PBMCs, DNA, RNA, and NGS including Whole Genome Sequencing (WGS).
329 RC patients have been accrued. We perform NGS testing depending on tumour purity (TP), including WGS from fresh tumour samples (and on a select number of FFPE cases). We have performed WGS on 88 samples, of which 77 (69 fresh frozen, 8 FFPE samples) were rare gynaecological cancers. H+E tissue sections from biopsies or surgical resections were reviewed by our anatomical pathologist for morphology, diagnosis and TP. We noted a 13% failure rate (4/30 samples) from the first third of the WGS cohort (“insufficient tumour for somatic variant analysis”), with no QC point along the processing pathway.
A minimum of 20% tumour cellularity was required for WGS. Thereafter, to avoid the risk of sample processing failure, fresh frozen samples were only submitted whole if the entire sample TP was >40%. If TP<40%, tumour areas were microdissected directly for DNA extraction. We report on 16 of the subsequent 52 cases which required microdissection. Of these, two are pending and two failed despite microdissection ie two failed out of 52 cases, 3.8% fail rate for WGS completion. Zero out of 36 cases remaining, failed WGS.
Of the successful 12 microdissected cases, actionable aberrations were detected in nine cases, five of which were highly actionable. We consider that high risk of failure of rare cancer samples unacceptable; microdissection can reduce this risk from ~13% (4/30 overall samples) to 3.8% (2/52 cases). Patients with rare cancers deserve the best possible therapeutic guidance; when in a position to have WGS, they should have optimal preparation of their sample.