Skip to main content
Premium Trial:

Request an Annual Quote

VA Researchers Examine Off-Label Therapy Use in Precision Oncology Program

VA-veterans affairs

NEW YORK – Research from the Department of Veterans Affairs National Precision Oncology Program (NPOP) found about 1 percent of cancer patients treated in the program received an off-label targeted therapy for an actionable mutation.

In a study published last month in JCO Precision Oncology, VA researchers analyzed real-world data from 18,686 cancer patients who underwent comprehensive genomic profiling (CGP) through NPOP between February 2019 and December 2021. They hoped to better understand the use of off-label targeted therapies in the VA system and its impact on patient outcomes, according to lead author Vishal Vashistha, staff hematologist and oncologist at Raymond G. Murphy New Mexico VA Medical Center and corresponding author on the study.

NPOP began in 2016 for patients across the US being treated at Veterans Health Administration hospitals. The program provides patients access to CGP through external diagnostic firms and provides treating physicians with advice and access to matching targeted therapies. Any VA provider can request the testing and services of NPOP on behalf of their patient, according to a spokesperson for the VA, and NPOP pays for the testing.

The VA also specifies treatment pathways by cancer type, the spokesperson said. For example, lung cancer patients with stage IIa or higher disease can receive CGP or PD-L1 testing to determine tumor proportion score, or what percentage of tumor cells express PD-L1. For prostate cancer patients, those with metastatic disease can receive CGP and those with high-risk or metastatic disease can undergo germline genetic testing, according to the VA's clinical pathways.

In the recently published study, the researchers found off-label drugs were prescribed to only 0.9 percent of patients who underwent CGP, while 11 percent were prescribed a targeted therapy either on- or off-label.

Off-label treatments were prescribed across many cancer types, with the highest prescription rate among thyroid (8.6 percent), breast (7.6 percent), and small intestinal (3.3 percent) cancers. These treatments were also prescribed for variants in 31 genes, and the most common alterations for off-label treatment were BRCA1/BRCA2/ATM (22.5 percent), ERBB2 (19.5 percent), and BRAF (19.5 percent). Patients in the study had CGP using one of three Foundation Medicine panels: FoundationOne CDx, FoundationOne Liquid CDx, or FoundationOne Heme.

Going into the analysis, Vashistha said he expected the rate of off-label prescriptions to be higher than what they found.

"The reality is that even though comprehensive genomic profiling and next-generation sequencing is very exciting and it's led to a lot of discoveries about tumor biology and has opened some opportunities for some cancers, the actual utility is still quite low," he said. "So, it wasn't too surprising that the rate of off-label treatments was relatively low."

He also noted that physicians who want to prescribe off-label drugs need to make the case for its reimbursement, which can be a barrier to access.

"Even though the VA's insurance model is more flexible than private insurance and Medicare, if you're going to request a drug that doesn't have a clear indication for its use, you have to make a compelling argument to pharmacists, to the local pharmacy group who is about to agree to pay for drugs that are fairly expensive," he added.

He also noted that the three gene groups that had the highest rate of prescriptions is likely explained by the rate at which these mutations occur in patients and the number of targeted therapies approved for these mutations.

"The homologous recombination repair genes, BRCA1/2 and ATM, there is very exciting data in prostate and ovarian cancer showing patients respond to PARP inhibitors," he said. "ERBB2 has a number of different anti-HER2 therapies associated with it that do not carry significant toxicities at all. The comfortability with [Genentech's HER2 therapy] trastuzumab is so high from breast cancer that even I'm more comfortable prescribing that drug than I am a lot of chemotherapeutic agents."

He said that the rate of BRAF mutations is quite high across cancers, which probably led to the high rate of off-label prescribing. Vashistha also noted that the US Food and Drug Administration approved the combination of Novartis' Tafinlar (dabrafenib) plus Mekinist (trametinib) for any solid tumor harboring a BRAF V600E mutation shortly after their study completed.

Vashistha's team also explored the level of evidence required for off-label use, using OncoKB levels of evidence. The OncoKB levels range from 1, to denote an FDA-approved drug in this indication, to 4, which denotes hypothetical evidence such as biological evidence that supports the biomarker being predictive of response to a drug, without clinical evidence.

Of the 183 total prescribed off-label regimens, 13.1 percent carried level 2 evidence, which means it was for a standard care biomarker recommended in professional guidelines as being predictive of response to an FDA-approved drug in this indication. Nearly 4 percent of off-label regimens had level 3A evidence, denoting there is compelling clinical evidence that supports the biomarker being predictive of response; and 71 percent carried level 3B evidence, for a standard care or investigational biomarker predictive of response to an FDA-approved or investigational drug in another indication. Twelve percent of regimens had level 4 evidence, and three regimens were not associated with OncoKB levels of evidence.

According to the VA spokesperson, most targeted therapies for cancer are "non-formulary" and require prior approval requests.

"If requested, the NPOP will provide a recommendation from an interprofessional expert panel, which will typically consider recommending off-label use if supported by OncoKB level of evidence 2, 3A or 3B," the spokesperson said.

Many patients and oncologists have struggled to access off-label cancer treatments and molecular testing due to insurance coverage. However, molecular testing coverage has improved in recent years. In 2019, the Centers for Medicare and Medicaid Services agreed to cover the cost of next-generation sequencing for patients with advanced cancers for tests that are performed in CLIA-certified labs and have been approved or cleared by the FDA for use as a companion diagnostic. Off-label drug coverage varies widely by insurer, with many payors requiring similar appeals and prior approval before agreeing to cover.

However, the VA research and previous studies have shown that off-label drugs can lead to responses across cancer types. In the VA study, 26.5 percent of patients who received an off-label targeted therapy responded to the treatment. The researchers noted that those with level 4 OncoKB evidence had lower progression-free survival compared to those who received therapies with level 2, 3A, or 3B evidence.

Vashistha noted their findings were in line with previous studies examining the outcomes of patients with off-label treatment, such as the MyPathway basket trial that demonstrated a 23 percent response rate for patients across tumor types treated with off-label therapies.

The VA research also shows that off-label therapies and wide use of CGP is not restricted to academic medical centers, Vashistha said. NPOP facilitates CGP and its expert panel across all VA hospitals in the US that treat cancer patients.

"The findings clearly convey that we do a pretty good job of giving access to off-label therapies to our oncologists, provided that they have a gene variant of interest and they can make a compelling argument for why a drug should be prescribed," Vashistha said.

Beyond the VA, Vashistha emphasized that while it's relatively rare for a patient to have an actionable genetic variant, this study shows that off-label treatments can lead to responses for some patients.

"The response rate, the duration of response, the progression-free survival is dependent on the level of clinical evidence that is available [for the off-label therapy]," he said. "Making decisions based upon just biochemical rationale, or level 4 OncoKB evidence, clearly has lesser benefit than those that have clinical evidence for their treatment. It's a balance — one should consider prescribing off-label target therapies if they're able, provided that their insurer is in agreement and the patient is interested, but that decision should be based upon clinical evidence."