Review Reveals Challenges of NGS Testing in Community Practices, Proactive Measures to Increase Utility

Jennifer Marie Suga, MD, a medical oncologist at Kaiser Permanente Northern California

Jennifer Marie Suga, MD

The importance of next-generation sequencing (NGS) has become apparent in identifying genetic aberrations across tumor types, said Jennifer Marie Suga, MD. However, challenges regarding the utilization of genetic testing in the community setting remain, thereby limiting the number of patients who are able to benefit from NGS.

“Although NGS is a powerful tool, it is being used more in the academic setting, and less so in the community setting,” said Suga. “In the academic world, there could be additional benefits to [using NGS] as a powerful research tool to find clinical trial candidates and characterize the kind of cancer a patient has. In the community, [the utilization of NGS] is a different story.”

To better understand and remedy the impediments to NGS testing in the community setting, investigators conducted a review-based study evaluating the impact of NGS within Kaiser Permanente Northern California (KPNC).

Of 4825 patients who underwent NGS, 4.3% were identified as being potentially germline mutation carriers. Of these patients, more than half (54%) did not meet the 2020 National Comprehensive Cancer Network (NCCN) Criteria for Genetic/Familial High-Risk Assessment. Approximately 40% of these patients received subsequent genetic counseling and testing. A genetic alteration was confirmed in 16.8% of these patients, suggesting the need for more inclusive and enabling parameters around genetic testing in health care.

In an interview with OncLive, Suga, a medical oncologist at KPNC, discussed the importance of understanding the current limitations to NGS in oncology, the impact of the study findings, and remaining challenges regarding NGS testing in the community setting.

OncLive: How is NGS currently being used in the community setting?

Suga: NGS is a powerful tool that permits the characterization of [and identification of] mutations within hundreds of genes that are important for cancer [proliferation].

As our field has evolved, we have seen many more drugs that are able to target these mutations and elicit great responses. We have been able to try to identify these patients who may significantly benefit from these treatment options that are now coming out.

It is estimated that of all patients with advanced cancer, about 15% to 20% receive comprehensive genomic profiling such as NGS. Part of that low uptake of NGS, at least in the United States, is because a lot of [patient] care occurs in the community setting. Many community oncologists feel very uncomfortable with understanding and interpreting the results of NGS tests and may be hesitant to order the test.

What were the objectives of this study?

The study aimed to address how to increase the utilization of NGS in the community setting. A practical challenge is lack of access [in large] scale. Many institutions and health care systems don't have robust infrastructure and processes to be able to deliver routine, broad spectrum, NGS to all patients with advanced cancer. Our goal was to implement a comprehensive genomic oncology program that supports community oncologists from the time of ordering the tests, to receiving the results, to interpreting the test, so that they are able to increase the utilization of NGS.

To do that, we partnered with Strata Oncology, who, through their clinical trials protocol, provided a forum where we could provide NGS tests. Then, we worked to implement the whole process from ordering [the test] all the way to interpreting the results.

KPNC is an integrated health. [We at] KPNC felt that it was important to work toward increasing access to NGS and ensure it was a comprehensive program that would allow us to do that.

Why is multidisciplinary NGS assessment important?

We felt it was important to involve other disciplines beyond oncology to get a good sense of the support [that exists] for the treating oncologist. Our genomic oncology committee not only encompassed medical oncologists who specialize in specific diseases, but other specialties including pathology, genetics, and research. We have population researchers and clinical trial personnel who we can readily patients with for clinical trials.

What was the design of the study?

Once we designed our comprehensive approach to genomic treatment, in November 2017, we collected every NGS test that was ordered through our system through December 2019. We found that we were able to significantly increase the number of tests being ordered. We started off with a handful of tests per month prior to initiating this program. Then, we were able to reach approximately 300 NGS tests per month.

Then, we started streamlining the ordering system from the physician to the pathologists. We then allowed for central collection of the test results, which would be quickly transmitted to the treating physician through email and electronic medical records.

At the same time, we created a database so our genomic oncology committee could review every case. We aimed to help the treating oncologist in real-time regarding [how to manage] these patients.

What did the results show?

Our genomic oncology committee evaluated each of those results for several things. First was whether there is an FDA-approved drug for that particular patient. Second, was whether there was an FDA-approved drug for an alternative cancer that has clinical evidence [of activity] in the particular cancer [the patient had]. Third, was whether there was a clinical trial within our network or in our local area. Fourth was whether the patient required additional genetic testing and counseling as determined by our genetics [team].

Findings of this comprehensive review showed that approximately 40% of patients had an actionable mutation. Of those 40% of patients, certain tumor types like lung cancer and breast cancer have about a 50% rate of actionable mutations. About 22% of those patients were potentially eligible for a clinical trial within the area and about 11% were eligible for an FDA-approved drug or one that was promising in that indication.

In another related ASCO abstract that was presented by my colleague, Sachdev Thomas, MD, of KPNC, we showed that approximately 4.3% of [patients] we evaluated had a germline mutation. [In retrospect], over 50% of those patients would have never been referred to a geneticist based off of clinical characteristics and NCCN guidelines.

How might these results impact clinical practice?

The implications are likely twofold. One is understanding that even in the community setting, NGS tests are clinically meaningful to the patient's [treatment course] about 40% of the time. The other is that we were able to increase our rates of NGS testing by conducting a systematic comprehensive approach from helping to get the test ordered, quickly relaying the results to the patient, and helping with the interpretation of the results rather than relying on the treating oncologist [to do all of the above]. That [approach] increases the comfort level of oncologists ordering these tests.

Another important aspect to offering more access to NGS testing for our patients with advanced cancers has increased our opportunities to participate in genomic clinical trials research. Genomic oncology clinical trial offerings increased 3-fold in our institution prior to and after initiating our genomic oncology program, going from 4.4% of our total clinical trials to 11% of our clinical trials portfolio.​​​​​​​

What are some of the remaining challenges with NGS testing in the community setting?

NGS has some challenges, a lot of them being technical challenges. In the community setting especially, there is more reliance on small biopsies to diagnose cancer. It's often challenging to get [the] tissue [that] is required. As the technology evolves, more test results will be able to be given [with] smaller samples and will help to mitigate that. Perhaps, we will find ourselves [pursuing] liquid biopsies.

The turnaround time for NGS testing, which we know is really important, is specifically challenging in the community setting. By moving NGS up to diagnosis, we could provide important information up front and give patients an option to choose [optimal] treatments. It becomes very important to see if we can have the test results available at the time [of the initial consultation].

The turnaround time can be a couple of weeks, and oncologists may be nervous to give the results [to patients]. Patients can become extremely nervous as well. Additionally, if a patient [is symptomatic], one may not be able to wait for the test [results], and the patient may have to start treatment. That is a practical challenge that occurs in the community that may account for some of the low utilization of NGS.

Cost is another important aspect to mention because it is not insignificant. We have to consider this, especially if there are co-pays for the test that the patient may have to bear. The decision comes down to whether the results of the test are worth the financial burden on patients. As the technology improves, I believe the cost of NGS will continue to come down, and it may not be as much of an issue in the future.

Even within our network of 120 oncologists, there are oncologists who are comfortable ordering this test for every single one of their patients with advanced cancer. Conversely, there are oncologists who almost never order the test, even though they have access to it. I would like to understand what some of the motivations are regarding why an oncologist will decide it is important to order NGS, while another oncologist, who may be seeing the exact same number of patients, [thinks otherwise].

How could cascade testing and automated workflow allow for a greater utilization of NGS?

Cascade testing is a systematic approach to identifying individuals at risk for hereditary conditions. Having the geneticists involved extremely early in that process and flagging those patients who might have a germline mutation has significant implications for not only the patient, but also the patient's family members. We have learned that it is also important to consider [cascade testing] when you're implementing a large scale NGS platform.

In terms of automated workflow, we try very hard to incorporate as much automated workflow as possible to [streamline the process for oncologists and pathologists]. Reflex testing is going to continue to evolve as we automate workflow. [In this regard], at the time of diagnosis, patients would be automatically [tested] without having to have the oncologist ask for the test. We're working on some of those ideas to automate [NGS] further and make it easier to receive those tests.

How has this review aided our understanding of genetic testing in clinical practice?

From implementing this program, we've learned that if we are able to provide a comprehensive approach and a system that streamlines the entire process of NGS testing, oncologists [will want to order the test]. The reason why we're able to see a rapid uptick in the number of tests being ordered is because we are able to provide that support from beginning to end. As more oncology drugs continue to receive regulatory approval for rarer cancers, the clinical actionability rate with NGS will continue to increase. The challenge for the community is to determine how we can support the use of NGS for oncologists who have busy clinics. It is challenging to keep up with all the new indications and treatments that are coming out every day.

Thomas SP, Suga JM, Truong, TG, et al. The impact of tumor NGS testing on hereditary cancer risk assessment and population management in an integrated community health care system. J Clin Oncol. 2020;38(suppl 15; abstr 1517). doi:10.1200/JCO.2020.38.15_suppl.1517

<<< 2020 ASCO Virtual Scientific Program