Do nodal counts indicate quality of colon cancer care?

January 1, 2008

Complete evaluation of the lymph node basin after surgical resection for colon cancer is important for the accurate identification of nodal involvement and for the complete resection of disease.

Complete evaluation of the lymph node basin after surgical resection for colon cancer is important for the accurate identification of nodal involvement and for the complete resection of disease.

Recently, the National Quality Forum, in collaboration with ASCO and several other professional groups, endorsed a 12-node minimum as a consensus standard for hospital-based performance. Large private payers have also begun to incorporate this measure into their pay-for-performance program.

But are nodal counts useful indicators of the quality of cancer care? Are patients with fewer than 12 nodes evaluated receiving poor care? Some colorectal surgeons think the answer to these questions is "no" and argue that effort is being misspent to enforce a concept that remains unproven.

More is better

"There is ample evidence that more is better, but based on the strength of the evidence, it may be premature to say that if a surgeon or pathologist cannot achieve more, he or she should not be reimbursed," commented George Chang, MD, assistant professor of surgery, The University of Texas M.D. Anderson Cancer Center.

A number of observational studies have indeed found that the evaluation of an "adequate" number of lymph nodes is associated with increased survival. "There is debate . . . although not as much as it might appear, at least in this regard," he added.

Dr. Chang is principal author of a recent review of 17 studies that validated the findings linking number of evaluated nodes with survival (Chang GJ et al: J Natl Cancer Inst 99:433-441, 2007).

"We have seen that as the number of nodes recovered and dissected by the pathologist increases, survival is improved. There is very little debate about this on the individual patient level," he said.

For example, citing one of the higher-quality studies in this series, Dr. Chang noted that Intergroup 0089 found a 14% higher absolute 5-year overall survival for stage II patients with more than 20 negative lymph nodes examined, compared with 10 or fewer. Among patients with lymph node metastasis (stage III disease), 8-year absolute overall survival for N1 patients was 34% higher for those with more than 40 nodes evaluated vs those with 11 to 40 nodes evaluated.

Similarly, in the INTACC study, an increase in the number of nodes recovered was associated with absolute improvements of 8% in overall survival and 17% in relapse-free survival in patients with Dukes' B2 or B3 disease (although Dukes' C patients were not affected by number of nodes).

A good standard

One surgeon standing solidly behind the 12-node standard is Luca Stocchi, MD, of the Cleveland Clinic. "It's a good standard. I feel strongly it should not be less than 12, and there is a lot of evidence to support this," he told ONI.

Dr. Stocchi suggested that the real controversy is not whether 12 nodes is enough but whether more than 12 nodes might be better. He noted that some studies have shown that evaluation of 13 to 15 nodes is better for the patient than the 12-node minimum, and at least one study found that number of nodes evaluated was an independent prognostic factor.

However, at the Cleveland Clinic, Dr. Stocchi and his colleagues have found no advantage in stage II patients for harvesting more than 12 nodes.

"As a surgeon," he said, "if the pathologist evaluates less than 12 nodes, and this is rare, I do call the pathologist, especially if the nodes are negative."

Hospital perspective

So while for individual patients, it appears that "more is better," in looking at the question from a broader, hospital-based point of view, "things are not so simple," Dr. Chang said.

Agreeing on this point is Sandra L. Wong, MD, assistant professor of surgery, University of Michigan School of Medicine, Ann Arbor. In a recent analysis (Wong SL et al: JAMA 298:2149-2154, 2007), Dr. Wong and her colleagues assessed whether hospitals that examine more lymph nodes after resection have superior late survival rates.

This retrospective cohort study used information from the national Surveillance, Epidemiology, and End Results (SEER) Medicare-linked database of more than 30,000 patients with nonmetastatic colon cancer.

The study analyzed hospital lymph node examination rates by quartiles. Hospitals were ranked according to the proportion of their patients in whom 12 or more nodes were examined. Median number of nodes examined was 6 for the first quartile, 8 for the second, 10 for the third, and 13 for the fourth. Late survival rates were assessed for each hospital group, with adjustments for potentially confounding patient and clinician characteristics.

On an individual patient level, an increasing number of nodes was associated with improved survival. However, on the hospital level, facilities with the most patients in the 12+ category had no better survival rates than hospitals with the lowest proportion of patients in this recommended category.

Although the four hospital groups varied widely in number of nodes examined, they were equally likely to be diagnosed with node-positive tumors and had very similar overall unadjusted rates of adjuvant chemotherapy.

"Focusing on the recently endorsed measure-that 12 or more lymph nodes should be the benchmark for level of hospital quality-we found that number of nodes did not significantly influence staging, use of adjuvant chemotherapy, or patient survival," Dr. Wong said. "From reviews by Dr. Chang and others, it seems clear that patients who have at least 12 nodes examined have improved survival-for whatever undefined reason. The problem is when you try to enforce this as a measure of hospital quality."

Most centers 'not there yet'

The Wong analysis may be representative of how well this benchmark is being reached in the community, Dr. Chang said.

"The numbers are improving nationwide, but most centers are not there yet," he said. Twelve or more nodes are examined in 90% of patients treated at National Comprehensive Cancer Network hospitals, in 95% treated at M.D. Anderson, and in approximately 45% in the 2003 SEER database, he noted.

"But in Wong's analysis, for the broad aggregate of hospitals, the numbers are not as good. Even among the highest-quartile hospitals, about half the patients did not have 12 nodes examined," Dr. Chang said.

Dr. Chang maintains that benchmarks for nodal counts should not be set or discounted in the absence of evidence.

Commenting on the Wong analysis, Dr. Chang said, "Their data support previous findings that the number of nodes evaluated is indeed associated with improved survival for individuals . . . but the findings from the aggregate hospital level are no more conclusive than findings in prior studies, including my review. We have yet to prove that even with perfect surgery, perfect evaluation, and 20 evaluated nodes, survival will be improved for a population of patients. We don't have all the pieces of the equation to prove or to discount lymph nodes as a quality measure."

Dr. Wong added, "Many stakeholders in cancer care have rushed to make this a quality indicator. But it is not exactly clear that quality indicators based on evidence from observational studies are appropriate at this time. They have rushed to anoint these indicators without really looking at whether this makes a difference in the end."

Patient node number varies

Nancy Baxter, MD, assistant professor of surgery, St. Michael's Hospital, University of Toronto, who has conducted outcomes research in this area, agreed and noted that the association between number of nodes and survival was made in observational studies and not randomized trials.

"People developing these quality measures should be very aware of the quality of research behind these recommendations," she said.

The issue, she said, is that patients have a variable number of lymph nodes, "and we need to find the readily identifiable ones and evaluate them. We need to focus on processes that ensure that this happens vs requiring a specific number of nodes. There probably is a lower limit below which you do not adequately stage the patient, but it's probably not 12."

Where to put the emphasis

The surgeons interviewed by Oncology News International agreed that the time and effort being spent on developing, teaching, and enforcing this quality indicator might be better directed elsewhere.

According to Dr. Wong, this quality indicator was "years in the making," and implementation will require many resources on a national level.

"We wonder if these resources couldn't be better spent in another arena," she offered.

Dr. Baxter commented, "It is understandable that we want to improve quality, but before we select quality measures, it is important to understand what we are measuring and to not, in haste, bring in measures that may have a paradoxical effect."

Focusing on lymph node recovery, she said, detracts from other things that might be more effective in improving quality of care.

Dr. Chang agreed but added a cautionary note.

"In looking for perfect quality measures that could be hard to identify, we should not overlook opportunities for incrementally improving quality. In trying to make a grand slam, we may miss opportunities for base hits."

Other factors may affect outcomes

The responsibility for adequate nodal evaluation falls upon the surgeon and the pathologist, but patient and tumor factors-some of which are still unclear-are also involved in outcomes.

Dr. Baxter speculated that some biologic factor influencing node number and survival-possibly a host interaction, ie, immunologic response-may be important. Studies being conducted in France have shown that the immune response of tumors is highly correlated with survival and, in fact, is a more robust factor than standard histopathologic staging, she said.

At this point, however, only the practice of the surgeon and pathologist can be controlled. Surgeons who perform substandard resections and pathologists who do incomplete evaluations are compromising care, the specialists agreed.

"We do an operation based on anatomic boundaries. But one could argue that if these are breached, you will get fewer lymph nodes, and that the diligence of the pathologist makes a difference," Dr. Wong said.

She added that "the pathologists tell us that no matter how diligent they are, there can be patient issues regarding number of nodes. Some just have more than others. We think pathologists are good at finding the positive nodes, and for that matter you only need one positive node to put the patient into stage III, which dictates the need for chemotherapy. Working to find more may not make that much difference."

Dr. Chang called for all surgeons and pathologists to closely examine their own practices and outcomes, paying attention to the well-established principles of colon cancer care.

For the oncologic surgeon, this care includes adequate resection of an appropriate length of bowel, the malignant involvement, and the mesentery, including the origin of the primary vascular pedicle and accompanying lymphatics, which is the primary location for positive lymph nodes.

The message to medical oncologists, Dr. Chang continued, is not to discount the importance of the number of nodes evaluated, prognostically and therapeutically, for the individual patient with colon cancer.

When number of nodes seems small or inadequate, this should "spark a phone call" to the surgeon or the pathologist to determine whether nodes might have been overlooked or what the underlying factors might be.