Clinical Pathways in Practice at Roswell Park Cancer Center

Stephen Edge, MD, vice president for system quality and outcomes at the Roswell Park Comprehensive Cancer Center and the University of Buffalo in Buffalo, New York, spoke with the Journal of Clinical Pathways about his experience using clinical pathways for oncology care, including the benefits, challenges, and considerations for practices considering implementing pathways.

Transcript: 

Stephen Edge, MD: I'm Stephen Edge. I'm a breast surgeon and a health services researcher and the vice president of persistent quality and outcome at the Roswell Park Comprehensive Cancer Center and the University of Buffalo in Buffalo, New York.

Tell us a little bit about the process for implementing a clinical pathway at Roswell Park.

Dr Edge: About six years ago, we felt that it was important to use clinical pathways to document quality both at the main campus as well as our community network. We spent some time considering writing our own pathways and recognize that the job can be overwhelming and therefore decided to go with one of the commercial vendors. And we currently use the Elsevier ClinicalPath system, formerly known as Via Oncology. This pathway primarily looks at the medical oncology or systemic therapy decision point and collects detailed information about the patient presentation and then makes recommendations based on that. We've leveraged those pathways for other uses as we'll talk about in a few minutes.

The implementation process started with multidisciplinary discussions across multiple disease site teams as we considered writing our own pathways, and that engaged the providers. Since then, we have gone to the commercial product. There's been some barriers for implementation. The doctors don't necessarily like having to do the few extra clicks. Every time I say it only takes about one or two minutes to enter it, they roll their eyes at me. But in reality, it only takes about one or two minutes to enter the information for a clinical situation. But I'm a practicing doctor too. I don't like extra clicks.

We've done pretty well with them, particularly across our network where we have virtually 100% utilization among the doctors. I think the community doctors actually find them more useful than our specialty doctors at the main campus cancer center.

How are guideline updates or changes incorporated into an existing pathway? 

Dr Edge: Well, they don't come from Elsevier. They come from a committee that's impaneled by Elsevier, which is a panel of oncologists at the sites that use the pathways. And each panel has a community oncology co-chair and an academic oncology co-chair. If one looks at the pathways, I like to say they're mom and apple pie, there's no surprises in the pathways in terms of treatment recommendations. The committee meets quarterly and they're pretty good about getting updates into the pathways. If an oncologist at any one of the practicing sites feels that there needs to be something updated in the pathways, they can submit that for discussion at the meeting. They're asked to submit the evidence supporting that. At Roswell Park, we actually have an agreement with Elsevier that we can modify the most common pathways if we feel that it's necessary to do so for our practice. And that dates back again to our initial attempts to write our own pathways. So again, the pathways are primarily mom and apple pie. No surprises there.

What are the advantages to implementing pathways at an oncology practice?

Dr Edge: Well, first and foremost is it's used as a decision-support tree. And as I said, the community oncologist or the general oncologist finds them as a good, I like to say, backstop. In other words, oncology practice has gotten so complicated with so many rapid changes and so many different diseases. The practicing oncologist can have a hard time keeping up and this provides them a good backstop. It also helps the patients understand that they're getting what's the standard recommended treatment for a specific situation.

Secondly, we are able to monitor the quality of care both at our main campus and across the network. We've used that for documenting that for our payers, for other external stakeholders, and internally. When I say that, I like to stress that we have found that the care provided both in our community practices and our main campus generally are in line with pathways except in circumstances where there are good clinical reasons for not to be in line with the pathways. Many people use the term off pathway treatment for those people. People who are treated off pathway generally have good reasons for being off pathway. We find that about 85% of the time the doctor treats patients with the recommended treatment for the clinical situation. But there are clinical circumstances, the patient has comorbidities, they've already received the drugs that are recommended, there are new drugs available that those kind of reasons justify going off pathway.

We actually had a paper at the ASCA meeting a couple of years ago where we specifically reviewed, audited off pathway reasons and showed that... The title of the paper was actually Off Pathway: Good Care or Problem? And we concluded that it was generally good care. In virtually every case, there was a well-documented reason. And for the clinicians on the audits, a very simple one was one of our senior breast medical oncologists had a 65-year-old woman with a three centimeter ER positive, HER2 positive breast cancer for which the standard treatment would be neoadjuvant therapy with chemotherapy and HER2 drugs, most likely TCHP. But she had severe brittle diabetes and cardiac history, had an ejection fraction of 25%. And he simply said, "There's no way I can get this woman through chemotherapy and I'm going to treat her with anastrozole plus trastuzumab." And you can Monday morning quarterback and maybe come up with a different regimen, but a highly experienced medical oncologist said, "I have to treat her with something besides the on pathway regimen." That's a justified or valid reason for going off pathway. It's not bad care, it's good care.

What is the process for requesting or approving an off-pathway treatment? 

Dr Edge: There's a number of different ways that people around the country audit and work with off pathway care. Some practices do prospective review and have the pharmacist or the P&T committee has to review any off pathway care. Other centers do no review. They just look at the on pathway rate. We've been doing it in the middle. We do a post-hoc review of off pathway care. And for almost four and a half years, we reviewed every case of off pathway care. We found that virtually all of them have good reasons and we've now cut back on that and are doing selected review for complex cases, or particularly now we're reviewing cases where multidisciplinary care prior to initiation of treatment may impact outcome. So I'm thinking of the neoadjuvant therapy for pancreas cancer or esophageal cancer or the like. And we'll talk a little bit later about how we're using it to identify people for advanced therapies.

What has your experience been with the DLBCL pathway so far? 

Dr Edge: Well, one of the things that you bring up is that there's a new CAR-T product, and I couldn't pronounce the name if you asked me. There's a new CAR-T product for people who have a progressive or relapsed diffuse large B-cell lymphoma. And given our broad community network, we want to be sure that those people who are seen with relapsed or progressive diffuse large B-cell lymphoma are identified and considered for what is a therapy which actually significantly improves their chance of a good outcome.

So we actually will screen anytime a decision is put into the pathway program for diffuse large B-cell lymphoma, and if it's the first time it's been put in, we'll look at that chart about two to three months later, at which point they should be having their repeat PET scan and see if they have progressive disease, or if they've had relapse disease and they already had a prior treatment, we review that record immediately. If we see that they haven't been considered for this, we actually call the doctor and say, "Are you aware of this and would you like them to be considered for this treatment?" We certainly do not tell the doctor how to treat the patient. The treatment of the patient is between the doctor and the patient. But we make sure they're aware of this. And in those cases, many of the people that we've identified actually come to Roswell Park for a consult.

We do have one affiliate who's actually closer to Boston and New York City and if that patient wants to go to Boston or New York City for that therapy, we would encourage them. Our goal is to get the patient the right care. So we're actually monitoring that. We're doing the same thing with myeloma at this point. We intend to expand that to other diseases to identify people who may be eligible for advanced therapies at a tertiary center, particularly those treated across our network.

What are some of the considerations or challenges for practices around implementing clinical pathways?

Dr Edge: The main one is deciding that you're going to do it. It is a change in the practice. It does require a little bit of extra work for the provider, but the benefits really outweigh those barriers and problems, the benefits in terms of monitoring quality, being able to identify what your practice is, identifying people for clinical trials. We haven't talked about a couple of the other major advantages of pathways. Clinical trials that are available and open at your center are directly in the pathway, and in fact, before you can select the standard treatment option, you have to acknowledge that there's a clinical trial that might be available for this person and make a active step to say either, "Yes, I want this person fully screened for the trial." Or, "No, I don't want the person screened because..." And then you have to enter a reason and the reasons include: I know the person's not eligible or the clinical trial's not available at my site and they can't travel or the patient has specifically declined the protocol. But you have to take an active action on the trial before you can get to standard treatments.

Another advantage is you can actually see what your practice pattern is. Before we had the pathways, even at our main center with Roswell Park, with all the resources we have, if you'd asked us how many people did you treat with metastatic HER2+ breast cancer, we couldn't do it. We couldn't pull those out of our records. So now if an investigator asks us for that when they're trying to design a clinical trial or submit a grant, we can usually tell them within 10 or 15 minutes the number of people treated annually with those clinical situations, something we never could do before. And that's being used widely by our investigators. So there's many advantages, but at the end of the day, I think the primary reason is to assure that people are getting the appropriate treatment, are being offered the appropriate treatment, and that you can document that.

I do like to say, I don't mean to be harsh, but at some point as a surgeon or a medical oncologist or radiation oncologist is about to initiate a treatment that's going to be life-changing, potentially life-threatening for a patient and systemic therapy may cost upwards of 100 to $200,000, I don't have a problem with asking to spend an extra two or three minutes to document why they're treating them the way that they're treating them, particularly if they're going outside the normal standard treatments. I think we owe it to our patients.
Interestingly, the American Society of Clinical Oncology and the Community Oncology Alliance have jointly developed a program for a patient-centered care in oncology practices, which they're calling their Oncology Medical Home. And one of the standards that they require for the Oncology Medical Home is the use of the pathways program, documentation of the treatment, importance of pathways, or specifically documentation of the reasons why people go off pathway. So this is actually being widely accepted in oncology practice.

What's the future though? The future has to move away from this very unidimensional, what is your systemic therapy for cancer? We will certainly be moving into multidisciplinary treatments. We will probably be moving into, I think it's an overused term, but artificial or machine learning–based systems that will help identify the clinical situation, collect all the information from the records, and potentially present that to the provider with treatment recommendations without requiring any further data entry. I think we're moving into really a fascinating time in medicine where we're going to have these kind of tools that we can really help us and ensure that patients get the right treatments.

Actually, that's going to really play out in a very important way as we more and more get into personalized therapies with specific targets based on genomic profiling of cancer. So the ability for an individual provider to remember and search all these treatment options is going to rapidly be drowned out by the volume of information and the availability of new therapies. It's going to be a really exciting time. In some form of pathways or decision support or algorithm, a tool is going to be available. But again, we have to move beyond these single decision point of what treatment you're giving for systemic therapy for cancer in our pathways. And we have to move beyond the provider entering these data, even though it only takes a minute or two.