Beakers in a research lab

A Seismic Shift: Expanding the Reach of Precision Cancer Medicine


Part of a Special Report by Duke Cancer Institute & the Department of Pathology, Duke University School of Medicine — as featured in the 2021-22 Department of Pathology Annual Report (pdf)

Oncologists today have a wider range of anti-cancer drugs to reach for, many of which target the molecular alterations believed to contribute to the cancer’s development.

Comprehensive genomic profiling, also known as next-generation sequencing (NGS), is used to identify these molecular alterations. Duke Cancer Institute (DCI) oncologists partner with Duke University Health System (DUHS) Clinical Labs and private diagnostics companies to test patients at diagnosis and/or after the cancer grows or spreads.

While it can vary across cancer types, increasingly, targeted therapies that can save patients from needing toxic chemotherapy are becoming available at multiple points in a patient’s cancer treatment, from first line standard of care to subsequent treatment after progression on conventional therapies.

Test results are entered into a Molecular Registry of Tumors known as Frameshift MRT. This centralized informatics tool — designed, built, and coded at Duke by Michael Datto, MD, PhD, (currently the medical director of DUHS Clinical Labs and vice chair for Clinical Pathology) and Christopher Hubbard (DUHS clinical informatics architect)  — helps oncologists identify if anything in their patient’s mutational profile, even extremely rare targets, can be treated with any existing targeted therapies or immunotherapies.

Duke Cancer Institute has been offering its patients NGS testing since 2014. Developing Frameshift MRT three years later to organize and optimize the growing volume and complexity of data, and the subsequent formation, in early 2018, of a weekly multidisciplinary Molecular Tumor Board to review complex patient cases was a perfectly timed great leap forward.

The Precision Cancer Medicine Initiative — launched in 2017 by DCI, the BioRepository & Precision Pathology Center (BRPC), and the Clinical Labs — was the critical push behind it.

“It had become increasingly clear that the needs of sophisticated cancer researchers were changing across all cancer types; moving away from generic, archived, cancer-tissue samples, to fresh samples, to samples with a specific molecular abnormality,” explains Shannon McCall, MD, director of the BRPC, a DCI and Duke University School of Medicine Shared Resource housed in the Department of Pathology. “This coincided with clinical advances. Providers, including at DCI, were utilizing these broad molecular profiling assays to direct the care of cancer patients. There was a need to harness all this molecular profiling data to support both cancer research and treatment. I was totally on fire to get this started. We have so many big thinkers at Duke who said, ‘Let’s think about data and what’s possible.’”

In mid-2018, Executive Director of DCI Michael Kastan, MD, PhD, a noted cancer biologist, and Chair of the Department of Pathology Jiaoti Huang, MD, PhD, a prostate cancer researcher, signed a memorandum of understanding to co-fund the staffing necessary to further support the Molecular Tumor Board — co-directed by oncologists John Strickler, MD (for solid tumor cancers), and Matthew McKinney, MD (for blood cancers) — and to manage the Frameshift MRT database. This included hiring a bioinformatician/ data analyst (Jonathan Bell, PhD) and a savvy genetics scientist (Michelle Green, PhD).

Green, fresh from a position in the molecular diagnostic testing industry, joined the Duke Pathology (with salary support from DCI) in the spring of 2019 as senior research program leader of the Molecular Tumor Board and main user and manager of Frameshift MRT. She tracks promising clinical trials and new FDA drug approvals and has configured Frameshift MRT to automatically send therapy alerts to providers when their patients' molecular profiles match any known anti-cancer drug(s). This match could include drugs that are already FDA-approved, drugs that are “emerging” with strong clinical evidence, drugs that are being tested in clinical trials, or drugs that are approved or being trialed in another cancer type.

Over the course of the COVID-19 pandemic, Green has made several significant changes to Frameshift MRT that make it more user-friendly, interactive, and accessible for clinicians and researchers, who can access the Frameshift MRT dashboard when logged into the Duke VPN. Green is available to train and advise.

Marvelous Molecular Data Universe

As of August 2022, data from more than 9,000 tumor molecular testing reports, 8,000 patients, and 60 cancer types, as well as more than 5,700 hereditary testing reports are maintained in the Frameshift MRT database.

Mutations that can potentially be addressed with drug therapy have been found in more than half of endometrial, non–small cell lung, urothelial, breast, gastroesophageal, prostate, and bile duct cancer patients who have undergone NGS testing at Duke, as well as in 43% of colorectal and ovarian and 22% of pancreatic cancer patients. (data from April 2022)

More than 1,000 DCI patients have received more than one profiling test throughout the course of their disease; whether multiple NGS tests or a combination of both tumor sequencing and hereditary genetic testing. (When patients develop cancer-drug resistance — a common occurrence — they may be advised to have their cancer cells re-screened for new, potentially-druggable molecular targets).

“In the last decade there’s been a huge explosion in testing; these tests that they're doing now did not exist when I was in graduate school,” notes Green, who earned her PhD at Duke in 2011.

The “hit rate” of matching patients to targeted therapies and the number of patients having a dramatic response to treatment has continued to grow.

The accumulated tumor molecular and outcomes data captured for individual patients in Frameshift MRT over the past five years is also helping drive lab studies and clinical research promoting better cancer treatments.

Forty-five research projects and counting — including in colorectal, pancreatic, breast, and prostate cancers — have utilized Frameshift MRT.

“The magical part of Frameshift MRT is that it creates easy access to data for everyone. It works like Google, breaking down barriers," says Datto. "Dr. Green and the cancer precision medicine team invest an enormous amount of effort in data curation on the back end. The development team has layered on top of this a user-friendly interface that 'just works'. That makes Frameshift MRT much more than a database. It’s a simple way for Duke faculty and staff, learners, clinicians and investigators to engage with the universe of molecular cancer data at Duke. It’s like we strapped a giant ‘easy button’ on a highly curated, normalized vast data warehouse and this has revolutionized cancer care and research.


John Strickler and Bennett Caughey stand in front of Caughey's poster
Molecular Tumor Board co-leader John Strickler, MD, with precision oncology fellow Bennett Caughey, MD, at ASCO 2022. Caughey's poster earned him an ASCO Merit Award.

Actionable Mutations in Colorectal & Pancreatic Adenocarcinoma

Precision oncology fellow Bennett Caughey, MD, used molecular reports and matched clinical data from Frameshift MRT and the Japan-based SCRUM-Japan GOZILA and GI-SCREEN registry, to identify the optimal circulating tumor DNA shedding threshold (found in the blood) for detecting actionable (targetable) driver mutations in colorectal and pancreatic adenocarcinoma. Caughey collaborated with a group of researchers in Japan on the project, for which he won a 2022 American Society of Clinical Oncology (ASCO) Merit Award. The team submitted the study for publication and it's currently in review.

Caughey is in his second year as a precision oncology fellow with DCI. During medical school at the University of California San Diego (UCSD), he worked with Rafael Bejar, MD, PhD, to characterize the effects of mutations in genes regulating RNA splicing in MDS (myelodysplastic syndromes). In residency at the University of California San Francisco (UCSF), he worked with Eric Collisson, MD, to molecularly characterize a set of BRCA-mutated cholangiocarcinoma (bile duct cancer).

Immunotherapy & Prostate Cancer

Andrew Armstrong, MD, MSc, director of Research, DCI Center for Prostate & Urologic Cancers, and medical oncology fellow Landon Brown, MD (no longer at Duke) looked at metastatic prostate cancer cases driven by CDK12 mutations to gauge whether treating those patients off-label with the immunotherapy drug pembrolizumab was more effective in these patients than standard therapies. Metastatic prostate cancer cases that are driven by CDK12 mutations constitute an aggressive subtype with a high rate of metastases and short overall survival. These cases don't respond well to traditional endocrine-targeted, chemotherapy, or PARP inhibitor drugs.

The team collected retrospective data from seven centers, including from DCI (Frameshift MRT).

Study results, published in the Journal of Clinical Oncology-Precision Oncologyin 2020, showed immunotherapy to be effective in this subsetof patients,which was higher than what has been seen with similar forms of immunotherapy in unselected men with advanced prostate cancer. Another group published simultaneously on the same subject and found similar results. The clinical efficacy of immunotherapy (specifically pembrolizumab) is now being tested in the VA Office of Research and Development-sponsored CHOMP trial, which opened across many VA centers in 2019 and is still recruiting. (Rhonda Bitting, MD is the Durham VA site PI).

Current NCCN guidelines do not specifically recognize this patient population as a unique subset of patients in the treatment algorithm or recommend immunotherapy in this setting since the current data is all based on a small number of patients and is retrospective,” says Armstrong. “The hope is that the prospective clinical trial will validate this work and that of others, to then change our standard of care for these men.”

Immunotherapy & Breast Cancer

Duke Cancer Institute breast surgical oncologist Sarah Sammons, MD (now at Dana-Farber Cancer Institute) and colleagues at DCI, the Mayo Clinic, and Foundation Medicine, looked at deidentified national data, including from Frameshift MRT, on HR+ HER2– breast cancer cases characterized by a high tumor mutational burden (TMB) — cases where multiple molecular alterations of various types exist in one cancer patient — to investigate the efficacy of the immunotherapy pembrolizumab vs. standard-of-care therapy (endocrine therapy plus a  CDK4/6 inhibitor). In June 2020, the FDA had granted accelerated approval of the immunotherapy pembrolizumab to treat metastatic solid tumor cancers in patients with high TMB. The results and conclusions of this study were published in the Journal of Clinical Oncology-Precision Oncology in November 2022. The group found that while “clinical benefit to immune checkpoint inhibitors was observed in HR+ HER2–, APOBEC+, tumor mutational burden–high patients, further research is needed to optimize the treatment of APOBEC+ HR+ HER2– breast cancer and to investigate the efficacy of immunotherapeutic strategies in this population.”

Bridging the Gap: Equity in Testing

Since at least 2020, DCI has been working on ways to expand the reach of precision cancer medicine beyond its clinics and hospitals in Durham and Wake County to its Duke Cancer Network (DCN) affiliates, which are located primarily in rural areas of the Southeast, most in underserved communities.

Equitable access to quality care and clinical research is a core part of DCI’s mission as an NCI-Designated Comprehensive Cancer Center. The DCN is an extension of Duke into the community.

That’s why improved access to NGS testing, the integration of test data into Frameshift MRT, and clinical decision-making support from the Molecular Tumor Board, is now being extended, clinic by clinic, to DCN community care sites — beginning with four of five DCN locations in North Carolina served by DCI clinicians.

Earlier this year, Gibson Cancer Center in Lumberton, NC (Robeson County) and Scotland Cancer Treatment Center in Laurinburg, NC (Scotland County) became the first DCN sites to connect to these resources. Johnston Hematology/Oncology in Smithfield, NC (Johnston County) was recently added, and a fourth DCN affiliate, Maria Parham Cancer Center in Henderson, NC (Vance County), will soon join.

Three of the above sites are located in counties where the poverty rate is above 20% — Vance (21.3%), Robeson (26.6%), and Scotland (29.7%). Johnston County’s poverty rate is 11.4%. By comparison, the poverty rate in Durham County and Wake County — the location of DCI clinics and hospitals — is 11.7% and 7.4% respectively. (2020 U.S. Census)

A V-Foundation pilot grant (2020-2021) to a DCI clinical team led by Duke Cancer Network medical director Linda Sutton, MD — “Advancing Cancer Care in the Rural Southeast: Enhancing Precision Medicine and Institutional Collaboration in Community Cancer Centers” — kick-started this expansion process.

A team led by Datto built the underlying software infrastructure to support it, including a mechanism for integrating DCN patients' NGS testing results into Frameshift MRT.

Precision oncology fellow Bennett Caughey, MD — the first-ever precision oncology fellow at Duke — was brought on to help facilitate the workflow. He's now in his second year as MTB/Frameshift MRT liaison to the DCN sites. Caughey supports the Duke MTB in its educational and clinical mission in addition to pursuing clinical research in the field.

Hereditary genetics counselor Nicholette T. Sloat, CGC, MA, MS, is on the project team as well.

“There are definitely patients for whom molecular testing has been very relevant,” says Caughey, noting that more than 58 DCN patient cases of mainly lung, colorectal, prostate, and breast cancer have so far been reviewed by the Molecular Tumor Board.

This year the National Comprehensive Cancer Network Oncology Research Program (NCCN-ORP), with support from Eli Lilly and Company, awarded a two-year $300,000 grant to Strickler and team to pick up where the pilot grant left off, and specifically to "support the adoption of comprehensive genomic profiling (NGS testing) and biomarker-driven therapy selection for lung and thyroid cancer patients" at DCN sites. Caughey is the co-investigator and Green is the project manager on the grant.

“Therapeutic strategies targeting tumor-specific molecular alterations have shown dramatic and durable clinical benefit and revolutionized cancer treatment, particularly for lung and thyroid cancer patients,” says Strickler. "The use of comprehensive genomic profiling to guide drug selection for these cancers is rapidly expanding, driven by advances in next-generation sequencing technologies and several recent FDA approvals of targeted therapies."

Strickler describes “significant challenges to adopting these advances in the clinic.”

“Molecular Tumor Boards have been established at several academic institutions across the nation to provide physician support for the adoption of precision medicine, but these are not often available at community oncology centers.”

This means that patients may not be receiving guideline-based molecular profiling and appropriate targeted therapies. By extending MTB support and Frameshift MRT access to DCN oncologists, the team hopes to show a clinical benefit for patients.

With the grant, which takes effect this month, the team will track the rate of NGS tests ordered for lung and thyroid cancers at DCN sites compared to peer centers, the frequency at which actionable (druggable) molecular alterations are detected in these cancers, and the rate at which patients receive molecularly targeted therapies compared to peer centers.

As PI, Strickler will oversee all aspects of this project including planning, coordination, data collection, data analysis, and dissemination. Caughey will be the primary clinical lead for the expanded MTB serving the DCN; reviewing genomic profiling results and providing clinically actionable feedback to DCN providers. In addition, he will oversee data collection for cases reviewed by the MTB for the DCN.

Green will work with the research team to integrate DCN NGS sequencing results as they are received, into the existing Frameshift MRT, facilitate coordination between various DCN sites and the DCI, and assist with data interpretation, including systematic clinical trial and therapy matching.

Sloat will, in the context of this project, review NGS sequencing results and make appropriate recommendations for follow-up on molecular findings that could be hereditary (germline).

The project team also includes a Duke Pathology-based regulatory coordinator, a programmer, a research nurse, and a biostatistician.

Duke Cancer Institute social and molecular cancer epidemiologist Tomi Akinyemiju, PhD, MS, meanwhile, is leading a study — “Implementing Evidence-Based Interventions to Enhance Equity in Oncology Genomic Testing” — to examine the key sociodemographic, healthcare access–related, and clinical factors driving differences in NGS testing use between Black and White cancer patients at Duke Cancer Network community sites and DCI clinics.

For this V-Foundation funded project — awarded this year with funds from the 2021 Victory Ride to Cure Cancer — her team has linked retrospective Frameshift MRT data capturing orders for NGS tests and Duke Tumor Registry data across multiple cancer types and is interviewing oncologists and patients about their experiences and preferences related to genomic testing.

Preliminary results of Akinyemiju’s study confirm, like other studies, that Black patients are less likely to receive NGS testing — a significant barrier to accessing potentially life-saving cancer treatment.

“We observed racial differences among cancer patients with stage 3-4 tumors, who are most consistently indicated for next-generation sequencing. Specifically, 18% of Black patients with stage 3-4 tumors received next-generation sequencing in 2019 compared to 26% of White patients,” says Akinyemiju.

According to the 2020 U.S. Census, Black or African Americans make up 35.9% of the population of Durham County and 21% of the population of Wake County — the counties where DCI clinics and hospitals are located. They represent 52% of the population of Vance, 40.5% of Wilson, 18.3% of Johnston, 23.2% of Robeson, and 39% of Scotland — the five NC counties where DCN sites served by DCI clinicians are located.

Research already shows that Black patients are more likely to die from breast, prostate, lung, and colorectal cancers than White patients. While the reasons for this are multi-factorial, argues Akinyemiju, not getting access to these newer and better treatments that match to a patient’s unique molecular profile could widen that gap.

“There is an urgent need for transformative strategies to address inequities in genomic testing among Black patients with cancer,” says Akinyemiju, who serves as associate director, Community Outreach, Engagement, and Equity at DCI. “If racial disparities in access to genomic testing are not addressed now, Black patients will continue to have unequal access to novel targeted therapies, leading to even larger differences in survival.”

This page was reviewed on 12/02/2022