A pilot study by the Translational Genomics Research Institute using whole-genome and whole-transcriptome sequencing to guide therapy for patients with advanced cancer has demonstrated that such an approach is clinically feasible, despite setbacks in timing and other logistics.
In addition, the hurdles in the early study, published in PLOS One last month, have informed important improvements to TGen's strategy as it moves forward to demonstrate the clinical and cost benefits of clinical sequencing for cancer patients after opening a CLIA-certified lab earlier this year.
"The point of the pilot was to get a start in using whole-genome sequencing to try to identify targets that might inform possibilities for treating advanced cancer for patients that have progressed on one or two prior therapies," Glen Weiss, the study's first author, told Clinical Sequencing News.
According to Weiss, the study — conducted with Scottsdale Healthcare and funded by the National Foundation for Cancer Research — provides a sober look at the challenges that groups like TGen and others faced in early efforts to develop clinical whole-genome sequencing protocols and must address as they now refine their approaches.
"Even with reports of sequencing done at other institutions, sometimes [a turnaround time of] two to three months is not really something you can say is ready for prime time," Weiss said. "With sequencing for a couple of genes, you can do five days, sometimes sooner. That's really the benchmark that you have to have with whole-genome sequencing when you are talking about patients with advanced cancer."
In the TGen pilot study, of nine patients who received WGS (six of whom also received whole-transcriptome sequencing) between October 2010 and February 2012, the group managed to identify potential therapeutic targets in all but one. However, only one patient — with adenosquamous carcinoma of the pancreas — actually received treatment based on the sequencing results.
This individual had mutations in KRAS and PIK3CA identified by the TGen sequencing approach, which suggested that combined targeting of MEK/ERK and PI3K/AKT might be more effective than targeting just one or the other mutation. Based on these results, the patient was enrolled in a phase I PI3K inhibitor and MEK inhibitor combination study after progressing on chemotherapy and had a short-lived response based on PET/CT imaging as well as significantly reduced pain.
The pilot — which was also intended to inform development of a workflow for this type of clinical sequencing program — highlighted several challenges for making clinical whole-genome sequencing beneficial to patients, mainly long delays in managing fresh biopsy samples and difficulties reporting and communicating NGS results, to out-of-state participants and their oncologists.
In the study, the TGen group performed 100-bp paired-end sequencing on the Illumina HiSeq 2000, generating over 19 billion total reads, with average coverage ranging from 17x to 71x.
The median time from biopsy to completion and final analysis of NGS data was 91 days in the pilot, with 46 days the fastest turnaround and 243 days the slowest. The longest outliers were caused by sequencing instrument problems or low tumor percentage in specimens, according to the study authors.
The first patients received only WGS, but over the course of the study, technology and workflow improved enough that the group was able to add transcriptome sequencing and whole-exome sequencing for patients recruited later, although these added analyses also added to the time required for sequencing and analysis.
Another timeliness hurdle, according to the authors, was the fact that the sequencing was not performed in a CLIA lab, so validation of targets by alternate testing was more difficult, especially where samples were transferred between different institutions and not enough tumor tissue was available for subsequent validation of potential targets.
For two patients, sequencing results became available only after their death, while others progressed too far to benefit from therapy. Only one, patient number nine, was able to receive targeted treatment based on sequencing results.
"We did have at least one patient expire while we were still waiting for results to come out," Weiss said. "Another issue was with CLIA validation — when we had a target identified we couldn't get it validated either because of the custody chain, or because the region had some issues [preventing further analysis]."
Now, several years from the period of the pilot, TGen is continuing to sequence advanced cancer patients, notably within a study it is conducting in collaboration with the Mayo Clinic Phoenix/Scottsdale, Arizona.
TGen opened a CLIA lab this spring, and has been working to clinically validate whole-genome, exome, and transcriptome sequencing protocols on the Illumina HiSeq 2500.
Whereas the institute's early pilot effort struggled to provide sequencing results within a timeframe that could actually benefit patients, TGen now has a goal turnaround time of 14 to18 days for cancer sequencing.
"Shifting to the HiSeq 2500, having a better and more streamlined analytical process, and putting the whole thing in CLIA — just those three things along has shaved off [turnaround time] significantly," John Carpten, TGen's deputy director of basic science, told Clinical Sequencing News.
"The first patient in this pilot was consented back in 2010, almost four years ago, so, clearly a lot has changed," he said. "There have been improvements both in sequencing technology — particularly as it relates to speed and throughput — and also advancements in terms of algorithms for primary and secondary analysis and improvements in development of more automated platforms for reporting. With those improvements, we have clearly seen a substantial reduction in the timing of the process."
"We are getting close to that 14-to-18-day goal," he added. "Currently we are at 18 to 21 days for a complete comprehensive analysis, not just a panel, with long-insert whole genome, whole exome, and RNA sequencing and full clinical interpretation."
Carpten said a paper discussing the results of TGen's advanced cancer sequencing effort with Mayo clinic is currently in review. The institute is also applying its protocol in a melanoma trial in collaboration with the Barbara Ann Karmanos Cancer Institute at Wayne State University in Detroit, which is being funded by Stand Up to Cancer and the Melanoma Research Alliance.