April 2011
Feature Story

Anne Paxton

Moving closer to its goal of having the most user-friendly cancer protocols possible, in December the College approved a new process for making revisions in the protocols. The CAP Cancer Committee wants pathologists to know that the required elements are not carved in stone, and there is room for pathologists to pitch their ideas for change.

“The College is refining its cancer protocols by seeking input from community practitioners about what should be required, as well as insisting on a certain level of scientific evidence before an element is considered man­dato­ry,” says Paul Valenstein, MD, chair of the CAP Council on Scientific Affairs and director of clinical microbiology at St. Joseph Mercy Hospital, Ann Arbor, Mich. “We are envisioning a process in which the pathology community challenges its own experts to create a final product that synthesizes all perspectives.”

Released in fall 2009, the protocols represent a consensus on cancer reporting that has won approval from the central groups in cancer care, including the American College of Surgeons and the Canadian Association of Pathologists. But the College has set up the Cancer Protocol Review Panel to consider revisions to the protocols.

“Our thinking is that the protocols will best evolve in an environment in which there is broad input that includes experts and practitioners in the field,” Dr. Valenstein says. “So whenever new items are to be required in the checklist, there will be a public comment period during which all pathologists can weigh in, and a defined procedure will be followed internally at CAP that considers all comments.”

“We’re trying to have a process for continuing to improve the protocols that involves the membership as well as experts,” he says. It’s not a “wiki,” he cautions. “Protocols still need to be vetted by experts. But it won’t be an entirely top-down process either.”

Dr. Valenstein, in fact, recently submitted three suggestions of his own for revising the thyroid gland protocol issued in October 2009, and he points to these as examples of how community pathologists can ask for changes.

One of them, for instance, concerns the requirement that the size of each thyroid lobe be included in every report. Dr. Valenstein writes: “Should these measurements be required? I am not aware they have any prognostic value, and they don’t seem to meet the strength-of-evidence threshold CAP has now established for requiring elements. I admit these measurements are not difficult to collect, but with the American College of Surgeons’ Commission on Cancer recommending that required elements be listed in the synoptic format, requiring these elements forces them into the synoptic section of a report. This can clutter up the report and cause readers to overlook more important information. Please consider making them optional.”

Another one of Dr. Valenstein’s suggestions is directed at the required description of second tumors (in multifocal tumors). “Should detailed information about a second tumor be required? The authors sensibly point out that the protocol is designed to describe the dominant tumor, and the presence/absence of multifocality is listed elsewhere. I doubt there is evidence that detailed delineation of second lesions improves management (although I must admit to not knowing for sure). Please consider making this optional.”

If the committee decides to adopt a suggested change, it would probably take about six months for it to be final, Dr. Valenstein says. “There is a lot of version control that has to be dealt with, and there are linkages with the information system vendors that embed these protocols in their products. I am interested in seeing how the editorial panel will respond to my suggestions, but I understand it will take time for any suggestion to find its way into an update.”

Under the internal process the CAP is using to refine its cancer protocols, upgrading an element from “optional” to “required” is considered a major revision, while downgrading an element from required to optional is considered a minor revision and must pass fewer hurdles.

A stronger standard of required evidence is the other component of the Cancer Committee’s new approach to soliciting input on whether an element should be required, he says. “The evidence hierarchy of the National Health and Medical Research Council is being used to help determine whether there is sufficient evidence to require that an element be mandated in cancer reports. In general, evidence of level III-2 or higher will be required before a data element is mandated in the future. This means a required element must be shown to be important in a randomized or pseudo-randomized trial, a prospective cohort study, or a comparative study that includes a concurrent control group.” The College is also creating an appellate function to sort through disagreements about evidence, “so one set of experts won’t necessarily have the final say,” Dr. Valenstein says.

In general, the protocols have been favorably received, especially among the cancer registrar community, says Mary K. Washington, MD, PhD, chair of the College’s Cancer Committee and director of gastrointestinal and hepatic pathology at Vanderbilt University Medical Center. “We’ve had great feedback from them.” However, she says pathologists’ response to the protocols is often mixed, in part because of the transition from paper to electronic format. “People appreciate having the guidance of the protocols, but the format for the paper version is actually often very different from electronic, where you have pull-down choices from a menu, and selecting one choice from the menu would influence what comes up as your next choice.”

Integration or embedding of the cancer protocols into LIS software is still “at the very early stages,” Dr. Washington notes, but will be happening as the next LIS versions start coming out. It will vary by vendor, but the the ideal solution would be a true synoptic format where items are entered as discrete data elements and would be retrievable that way. However, she says, the committee is very aware that not every pathologist is going to update their LIS. “The protocols have to remain user-friendly for folks using paper formatting, and we would welcome feedback on how that’s working.”

As part of its ongoing review process for the cancer protocols, the College is also conducting a “global checklist initiative,” convening meetings between representatives of the CAP, the Canadian Association of Pathologists, the Royal College of Pathologists (Great Britain), and the Royal College of Pathologists of Australasia. Participants will compare cancer protocols in their respective countries and identify differences in what they consider essential. “We hope to use this information to refine our standards, requiring only the information that is generally recognized as necessary for the management of the patient,” Dr. Valenstein says.

When that minimum data set is developed, some of the elements that are required in the CAP checklist may be downgraded to optional, Dr. Washington says. The cancers on which the international group is focusing initially are lung, endometrium, prostate, and melanoma. “These were chosen because we don’t think there will be much difference for those four, while there may be more differences for checklists for colon and breast.” The British colorectal pathologists, for example, are vocal about the need for additional data on tumor deposits in colorectal cancer before incorporating these findings into staging systems. In addition, she says, “We have just adopted the 7th edition of the American Joint Committee on Cancer’s Cancer Staging Manual in this country, and they are still using the 5th edition for colorectal cancer in England. So these differences are going to be very controversial and may be challenging to resolve.”

The international group will be examining the data in the published literature and assessing the quality of the evidence for each element if there are disagreements. “We want the checklist to have all the elements needed for staging patients and for oncologists to make clinical decisions, but what we don’t want creeping in are the individual research interests—the things somebody thinks would be interesting to know, but that are not needed for clinical decisionmaking currently,” Dr. Washington says. For instance, sometimes surgeons want to know the weight of a specimen. “That’s an example of a practice that differs from country to country. We might end up with that as an optional element.”

The CAP cancer protocols represent an international standard for cancer reporting, Dr. Valenstein says. “Experts who author the protocols have made their best efforts to identify the types of pathology information that are necessary to manage cancer patients and that belong in pathology reports. But experts do not always agree with one another, and community practitioners can bring a different perspective to decisions about what belongs in a cancer report.”

Now that the College is insisting on strength of evidence, and there’s a procedure for practitioners in the field to put in a request for change, “we hope to harness the energy of practicing pathologists to help us make decisions about what’s required and what should be optional,” he says. Pathologists can be assured their suggestions will count. “You can make a difference; you can really make a change in what’s required.”