Oncologists' Off-Label Prescribing Pits Access Against Reimbursement

Kate M. O'Rourke


October 15, 2018

As common—and costly—as off-label prescribing is in oncology, it is poorly understood. It is, by nature, hard to study.

Yet it is clear that the practice is changing. Skyrocketing drug costs, tightening reimbursement, and a growing awareness of the financial toxicity of oncology drugs are curbing traditional off-label drug use.

In the other direction, oncologists are pushing boundaries with new immunotherapies, prescribing off-label on the basis of a drug's activity in a different setting and, experts worry, supported with little high-quality data.

There are few studies quantifying this trend, but a recent analysis of 168 oncology practices in the United States, including more than 2000 oncologists, found that 18% of prescriptions for nivolumab or pembrolizumab were off-label.[1] And this off-label use is pulling patients away from clinical trials, according to investigators.

Reimbursement, Payers Becoming More Restrictive

According to Robin Zon, MD, vice president and senior partner of Michiana Hematology Oncology, in Mishawaka, Indiana, reimbursement models are affecting off-label use, and payers are becoming more restrictive. Most reimbursement models, said Zon, are based on clinical pathways or guidelines from the National Comprehensive Cancer Network (NCCN) and American Society of Clinical Oncology (ASCO).

"In my experience and from what I have been hearing across the country, payers are requiring that, in order for you to order any drug or combination of drugs, they have to be considered an NCCN category 1 or 2a treatment," said Zon. "Medicare will pay for NCCN category 1 and 2a, and most payers follow the NCCN guidelines. If there is a new drug that I want to try based on early phase 1 or phase 2 data, if it's not part of the NCCN guidelines as category 1 or 2a or it's not part of an approved clinical pathway, odds are that it is not going to be paid for. It doesn't mean that you necessarily can't use a drug off-label, but you have to justify with evidence why you want to use it and hope it gets reimbursed."

According to Zon, the current reimbursement environment has changed the ability of clinicians to prescribe off-label. "In the past, off-label prescribing of oncology drugs was much more common. Five years ago, if I could find a phase 1 or early phase 2 trial that said this drug A is going to work for this patient with a particular disease that is failing every therapy, I might be able to get paid. But now, with payer influences and restrictions [from] adhering to utilization management strategies, I don't think it is happening as often," said Zon. She doesn't believe that clinicians are prescribing immunotherapy off-label and off-guidelines in the community setting without first knowing reimbursement status, because the drugs are so pricey.

In the United States, off-label use of 10 common cancer drugs accounted for nearly $5 billion in costs in 2010, and the costs are probably much higher with the arrival of immunotherapy.[2] When off-label drugs are not covered by insurance or accessed through compassionate-use programs, both patients and oncologists may be harmed financially.

"Private oncologists pay for drugs upfront. There are some practices that have paid $10,000 to $20,000 for one infusion, thinking that insurance will reimburse them, and then they don't get paid. The patient says, 'I can't pay you, but I'll go on a payment plan of $100 a month.' The person who ends up financing the treatment is not the patient; it is the provider who bought the drug," said Zon. "The patient is ultimately responsible, but they typically don't have $20,000 in their back pocket, so they go on a payment plan." Obtaining prior authorization can help protect clinicians and patients, but it doesn't always guarantee payment.

Efforts to Increase Approved Access

Clinicians should be following evidence-based medicine, said Zon, and for that, they rely on the US Food and Drug Administration (FDA). "In the past 3 years, the FDA has done a phenomenal job of amping up approvals, so they have changed the need to do off-label," she said.

The FDA approved 11 oncology drugs in 2016, 16 oncology drugs in 2017, and the agency was on track to break another record in 2018.[3,4] The FDA has been increasing its pace of drug approvals for novel cancer therapeutics, including anti-PD-1 agents, and greater than 60% of patients eligible for treatment with PD-1 inhibitors have them within 4 months after FDA approval.[5,6]

Don Dizon, MD, head of women's cancers at the Lifespan Cancer Institute in Providence, Rhode Island, and director of medical oncology at Rhode Island Hospital, believes that the tipping point—when insurance companies and clinicians started becoming more cost-conscious—was the approval of bevacizumab. This antiangiogenic therapy was approved in 2004 with much hype and a hefty price tag. Today, said Dizon, many institutions have systems in place that predict when an off-label request will be denied, and they can prepare for a denial by working closely with clinicians to compile relevant materials in case an appeal becomes necessary, even as the clinician is writing the prescription. Lifespan Cancer Institute has a system that determines whether there is a high or low likelihood of a drug being approved in an individual patient. These systems, including the one at Lifespan, can only function if doctors actively collaborate on appeals.

"Providers are giving a lot more thought about whether they want to pursue a drug that is not on-label, and institutions are requiring much more input from their physicians to prepare appeals before the drug is denied," said Dizon. "Although more work is needed, there is definitely a greater awareness of financial toxicity for our patients, and a lot of work has gone into—and continues to go into —sensitizing physicians about financial toxicity."

Off-Label Prescribing Trends With Immunotherapies

Many clinicians, said Dizon, want to ensure that all patients who are eligible for new agents can access them. "There is a lot of temptation from the oncologist's perspective to use newer agents, especially the checkpoint inhibitors, even if they are off-label; but it is forcing oncologists to do much more research on their own time," said Dizon. "There is much more work on the physician's part to get a drug to a patient off-label. It is not as easy as it was 5 years ago."

Oncologists have enthusiastically embraced immunotherapy even when the data aren't there.

"Within the gynecologic cancers realm, clinicians are more likely to prescribe immunotherapy for the diseases we treat, even it is off-label, such as for the treatment of relapsed ovarian cancer, where we may have exhausted standard treatments, largely based on the experience of immunotherapy in other treatments, such as renal cell and lung cancer, as well as melanoma," said Dizon. Immunotherapy, he said, might be used more often in rare tumor types and non-chemotherapy-responsive tumors, such as neuroendocrine carcinomas. "Technically speaking, it is off-label if I try to use lutetium (177Lu) oxodotreotide for an ovarian cancer neuroendocrine tumor that is somatostatin-positive, but in my opinion, it is not that far off from the label," said Dizon. The drug is a nuclear medicine treatment that is FDA-approved for somatostatin receptor–positive gastroenteropancreatic neuroendocrine tumors, but these tumors can occur elsewhere.

There is a great temptation, said Dizon, to use drugs that have shown remarkable success in individual cases of tumor types where the drug may not be approved. And he worries that some immunotherapy prescribing off-label is being done without the highest-quality data—ie, not from randomized trials. "We need to get people on clinical trials, especially if we want to find data and evidence. In fact, it's my strong preference to steer patients towards an appropriate clinical trial rather than off-label use of drug," said Dizon. "I do worry that clinical oncologists have enthusiastically embraced immunotherapy even when the data aren't there, and that this translated to the sense that [every] patient should be afforded immunotherapy, especially if they are at the point of progressing on standard treatment. For some diseases, such as ovarian cancer, whether immunotherapy has a role is still a big question."

While clinicians are more aware of financial toxicities, more physician education is needed, and more research into financial toxicity itself is needed. "In the course of recurrent disease, we have learned that financial toxicity is a negative prognostic indicator, but whether that is because patients can't access drugs because of cost, or whether they are abandoning other things that they require in their lives to afford the drug, is unclear," said Dizon.

Cathy Eng, MD, a professor of gastrointestinal medical oncology at the University of Texas MD Anderson Cancer Center, Houston, said she has seen an increasing use of immunotherapy off-label in recent years, both in the community and the academic setting. "Some of my patients came from the community where they were receiving an off-label immunotherapy agent based upon our earlier publication for rare metastatic anal carcinoma, because there is no FDA-approved drug in that setting and our study was positive," she said. Nivolumab provided an option for squamous cell carcinoma of the anal canal.[7]

Some patients may receive an off-label drug through compassionate-use programs, which requires a lot of paperwork, while others have the financial resources to pay out-of-pocket. "I have talked with colleagues who are performing molecular marker analysis on patients and, based on a generalized protocol at their institution, they were able to obtain drugs directly from the pharma company, based upon the patient's molecular markers, even though there was no specific trial in that setting," said Eng.

The Effect of Off-Label Prescribing

Off-label prescribing for patients can be beneficial, said Eng, but it prohibits people from seeking out clinical trials that are actively ongoing and reduces a patient's ability to participate in future clinical trials. Off-label prescribing also provides another line of therapy, which results in a more refractory population.

"The longer patients are pretreated, the more likely their performance status declines and the disease advances, and they don't have the performance status to be eligible for the trial," said Eng. "If patients receive the drug off-label and want to participate in a trial that uses that same class of drugs, they are ineligible. In the United States, we do a very poor job of enrolling patients on a clinical trial versus Europe or the UK, [where they are] very good about enrolling patients to clinical trials because they don't have off-label use."

Eng pointed out that off-label use of immunotherapies may increase risks for patients, unless physicians are familiar with the drugs, which carry uncommon side effects. Checkpoint inhibition is associated with a unique spectrum of side effects—immune-related adverse events—that fall into four main categories: gastrointestinal, hepatic, endocrine, and dermatologic.[8,9] Rash, colitis, enterocolitis, endocrinopathies, and hepatitis are common in patients on immunotherapies.[8,9,10,11,12]

Ed Kim, MD, chair of the department of solid tumor oncology at Levine Cancer Institute, Charlotte, North Carolina, pointed out that off-label prescribing is a spectrum that includes off-disease, off-line, off-dose, and off-formulation. "I don't have any statistics that off-label prescribing is worse now than in the past, but I think there is more attention to it now because drugs are more expensive than in the past," said Kim.

This is not a good way to utilize these agents.

Kim pointed out that for instance, 20 years ago, oncologists were using sequential chemotherapies for lung cancer, as well as other tumor types, which would have been considered off-label. "There was no FDA-approved chemotherapy for third-line treatment of lung cancer, but even the NCCN guidelines listed chemotherapies for third-line treatment of lung cancer. It wasn't uncommon for patients to receive three or four lines of chemotherapy for lung cancer," said Kim. "In that scenario, if you gave gemcitabine third-line or fourth-line for a patient with lung cancer, there wasn't a lot of scrutiny on that, even though there were no randomized data proving efficacy in that setting. Vinorelbine is still on the NCCN guidelines."

Treatment has evolved over time, and in addition to disease classification and organ classification, there is now molecular classification and testing of disease. Kim pointed out that imatinib (Gleevec, Novartis) and trastuzumab (Herceptin, Genentech) are FDA-approved specifically for certain tumors with molecular classification—c-KIT in the case of imatinib and HER2 overexpression in the case of trastuzumab—but clinicians can consider using the drug in a lung cancer patient whose genomic testing identifies a HER2 overexpression or a c-KIT mutation.

"If you are used to giving gemcitabine or vinorelbine as 'off-label therapy' in the third-line setting of lung cancer and now you have information that a tumor has a c-KIT mutation, you may think you are being smarter if you use dasatinib or imatinib rather than gemcitabine, because you are targeting a genomic variant where, in one tumor type, GIST, you have seen a profound effect," said Kim. "The scrutiny is that drugs are more expensive. Instead of costing a couple hundred dollars to give gemcitabine or $100 for [vinorelbine], you are talking about $30,000 for nivolumab or pembrolizumab. The higher cost associated with off-label prescribing is what is forcing us to look more closely at what is the true value for the patient. This is not a good way to utilize these agents, as we don't have much data to suggest efficacy with off-label targeted agents."

He said that some patients can obtain drugs from compassionate-use programs. "You have to be very cognizant of the cost to the patient," said Kim. He believes that if you are ordering whole genome testing on a patient's tumor, you need to have a study protocol open for access to drugs. "I don't support whole genome testing and then utilizing off-label strategies to treat patients," said Kim. "This should be done within the context of a clinical trial like ASCO TAPUR or NCI-MATCH, for example, where you can test the hypothesis and get an answer," said Kim, who is also chair of the TAPUR steering committee. "If you do it anecdotally or à la carte and don't track the data, you don't learn anything."

TAPUR (Targeted Agent and Profiling Utilization Registry Study) is a nonrandomized clinical trial that is working to describe the performance of FDA-approved, targeted anticancer drugs prescribed for advanced cancer that have a potentially actionable genomic alteration. The study provides FDA-approved targeted therapies contributed by collaborating pharmaceutical companies, catalogues the choice of genomic profiling test by clinical oncologists, and aims to identify potential signals of drug activity. TAPUR is open and enrolling patients at 113 clinical sites.

Another new avenue for patients to access drugs is the NCI-MATCH (Molecular Analysis for Therapy Choice) trial, where patients are assigned to treatment on the basis of the genetic changes found in their tumors through genomic sequencing and other tests. Several treatment arms are open to patients at any given time, each one enrolling patients whose tumors have a specific mutation. Most treatment arms will enroll 35 patients, but others will have double that. Drugs included in the trial either have been approved by the FDA for another cancer or are still being tested in other clinical trials, but have shown some effectiveness against tumors with a particular genetic mutation. NCI-MATCH also covers the drugs for patients.

Kim has been leading an effort over the past 4 years to modernize eligibility criteria for clinical trials, and he says this would decrease off-label use. "It is easy for me to say we have studies open, but if they exclude 90% of the patients that come to your clinic, it is not a practical study," said Kim. "We need to ease up on eligibility criteria so that patients on studies are more representative of the real-world population."

Kim, like most oncologists, prefers to see a patient on a clinical trial rather than be prescribed a drug off-label. "I prefer to have clinical trials available, but if you make clinical trials restrictive and you don't make them highly accessible, then you are actually forcing off-label prescribing," said Kim. "I believe that the solutions are more clinical trials that are accessible to underserved populations and liberalizing eligibility criteria to reflect real-world populations."

Trends in Off-Label Prescribing

According to Robert Mancini, PharmD, a clinical oncology pharmacist at St. Luke's Mountain States Tumor Institute (MSTI) in Boise, Idaho, off-label use of oncology drugs is more common with older agents. "If it's an older medication, especially one that has gone generic, it is more common to see off-label prescribing. The studies supporting use are out there, but no one is going to put the time or effort to get that additional FDA indication," said Mancini.

Mancini echoed other interviewees in saying that off-label prescribing is more common in the academic setting. "Academics tend to have more resources available to make off-label use happen," he said.

We have to be cautious about the false hope we are giving to patients.

Mancini believes that there has been an increase in off-label use of cancer drugs in the past decade. "A lot of that has to do with the increase over the years in the number of new drugs approved and how they are getting approved," said Mancini. "A lot of the drugs come into the oncology market in the relapsed- or refractory-line setting and people are trying to use it earlier and earlier."

Data from studies appear to back this. In a study of 2663 women with breast cancer treated in the United States between 2000 and 2009, 13% of the drugs were prescribed off-label.[13] A study evaluating first-fill oral chemotherapy prescriptions between 2011 and 2013 at MSTI estimated that 30% of the oral chemotherapies were prescribed off-label; roughly 90% of the off-label use was supported by NCCN guideline recommendations.[14] The top five off-label oral chemotherapies prescribed were capecitabine, temozolomide, lenalidomide, abiraterone, and everolimus.[14] Other recent studies estimate off-label use at 30%.[15,16,17]

Off-label use includes disease off-label, setting off-label, and line of therapy off-label, said Mancini, and much of off-label prescribing today is using something frontline that is not approved for frontline use. He said he is seeing an increasing number of situations where clinicians are targeting genomic markers with off-label use, even though it is not FDA-approved for that. "There are a lot of diseases where we see BRAF mutations, but if you look at the indications for the BRAF inhibitors, they don't include the particular disease state," said Mancini.

Mancini says there is also an increase in off-label immunotherapy with the PD-1 and PD-L1 inhibitors. "There are a lot of studies in a lot of different diseases states, and some of them have captured the indication and some not yet, so I think there are some options there," said Mancini. "If you have a good phase 3 study in the literature, some insurances will take that into account. Relying on phase 2 is a bit more sketchy in terms of likelihood of acceptance."

Mancini believes that PD-1 and PD-L1 inhibitors are used off-label more often for non-Hodgkin lymphoma because the drugs have the indication for Hodgkin's, and for pancreatic cancer because there are new data that haven't been put into the package inserts. Off-label use supported by evidence and guidelines is good for patients as long as patients have realistic expectations. "Until the data are more mature, we have to be cautious about the false hope we are giving to patients," said Mancini.

Echoing others, he said that insurance coverage—not data—is what is driving the ability to prescribe off-label. "Every insurance has their own qualifications on what they will allow or pay for off-label," said Mancini. He said clinicians should do all they can to communicate with patients. "I have seen situations where I am about to start this treatment and there was no consideration as to whether or not the insurance would actually pay for it and what that means," said Mancini. "In the end, the patient thinks they are getting this treatment, but their insurance can't pay for it and they can't pay because the costs are so high, so it's not something they pursue."

Drs Eng, Kim, Mancini, and Zon have disclosed no relevant financial relationships. Dr Dizon disclosed that he serves on the data safety monitoring boards for clinical trials involving AstraZeneca and Regeneron Pharmaceuticals, and his institution receives clinical trial funding from Bristol-Myers Squibb.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.