NEW YORK CITY—Despite a number of studies looking at surveillance strategies in colorectal cancer, the best form of follow-up for patients who have undergone tumor resection is still heavily in question. While select patients may be cured following surgery, a fairly high rate of disease recurrence still remains, magnifying the need to determine better models of surveillance for this at-risk population.
At the Great Debates and Updates in GI Malignancies conference, two leaders in the research and treatment of colorectal cancers examined the existing literature and argued for either a high-intensity or low-intensity method of follow up, taking into account the benefits, harms, and cost implications of each strategy.
A concept that is critically important in surveillance, said Al B. Benson III, MD, is not only figuring out which patients will likely have disease recurrence, but when.
Citing data presented by Daniel Sargent, PhD, at the most recent Gastrointestinal Cancers Symposium, Dr. Benson noted that 85% of recurrences occur within the first three years following resection. “Fine tuning of these data, looking at overall and disease-free survival, particularly for Stage III patients, there's no question that chemotherapy reduces the risk of recurrence,” he said. “And when it does, it is maintained over time.” However, the impact of chemotherapy is seen only within the first two years, after which there is no benefit.
In a 2003 meta-analysis by Jane Figueiredo, PhD, et al looking at six randomized trials comparing intensive versus conventional follow-up, in which intensive follow-up included the use of scanning, the results favored the more intensive strategy in terms of identifying patients whose disease recurred and the percentage of people who were able to go to surgery—suggesting an impact on survival, Dr. Benson said.
Furthermore, a subanalysis of the data looking at mortality at five years among patients who did not receive carcinoembryonic antigen (CEA) testing vs those who did showed that CEA testing had a more favorable outcome on overall survival (overall relative risk ratio of 0.71 vs 0.90). A subgroup analysis of CT scan vs none also favored imaging in terms of the impact on survival (overall relative risk ratio of 0.74 vs 0.91), he said.
A more recent trial led by Francisco Rodriguez-Moranta, MD, looked at intensive vs simple surveillance, which was a physical CEA test and colonoscopy, and concluded that Stage II patients reap the greatest benefit from intensive surveillance.
“That makes some degree of sense, because we know in terms of liver resection, for example, that it is somewhat stage dependent based on the primary tumor, so Stage II patients who recur tend to do better than Stage III patients who recur,” Dr. Benson said.
Another large, Intergroup, randomized laparoscopic colectomy clinical trial by Vassiliki Liana Tsikitis, MD, and colleagues showed that both early-stage and late-stage patients fair equally well in terms of surveillance strategy and patterns of recurrence. Those authors did note, though, that later-stage patients, particularly Stage III, are more likely to have multiple sites of disease, with an impact on survival.
“When [these authors] compared methods of detection and first recurrence, there really was no difference, and it appeared that the dominant way to detect recurrence was CEA and CT scan,” Dr. Benson said.
The most robust analysis evaluating surveillance is the 2007 Cochrane Systematic Review of Follow-up Strategies for Patients Treated for Non-metastatic Colorectal Cancer, Dr. Benson continued. Based on the studies included in the meta-analysis, the Cochrane Colorectal Cancer Group concluded that intensive follow-up is an appropriate strategy because recurrences will be detected earlier, which will have an impact on survival.
Because of these data, guidelines from both the American Society of Clinical Oncology and the National Comprehensive Cancer Network not only emphasize the importance of risk assessment, but also favor a more intensive strategy, particularly during the first one to three years, Dr. Benson added.
“In general, our recommendations include assess and discuss risk with the patient in terms of recurrence and a second primary—this is where colonoscopy becomes so important.
“The greatest benefit is clearly for patients who have resectable metastases. The greatest risk for relapse would be Stage IIb, IIIb, and IIIc patients, and the tools we now have include the physician visits for coordination, with the utilization of a strategy of CEA, CT scanning, and colonoscopy.”
David P. Ryan, MD, said that the first question that needs to be answered when discussing surveillance is whether there is any evidence that earlier vs later treatment affects survival in multi-site disease.
“If your average patient presents with multi-site disease, do you really care about starting FOLFOX Avastin or FOLFIRI Avastin six months earlier or a year earlier than you otherwise would have found it? No,” he answered.
Therefore, intensive surveillance is applicable only in single-site disease. “We're certainly not going to help people feel better by intensive surveillance, we're not going to make them live longer, but we might cure them if we pick up isolated metastatic disease earlier than we otherwise would. We also want to find new colorectal primaries so we can do colonoscopies, but really, how many people does this apply to?”
Using the Cochrane Database of Systematic Reviews as well, Dr. Ryan noted that only about 20% of patients are eligible for resection of isolated metastatic disease, with another potential 10% being able to be converted from having unresectable disease to resectable.
The most compelling argument in the Cochrane analysis is the fact that people who received some form of liver imaging did better than those who did not, Dr. Ryan continued. However, the data supporting improvements in mortality and recurrences are not as compelling when examining the database in greater detail.
“Meta-analyses are only as good as the data that go into them,” he said, noting the range of definitions for low-intensity and high-intensity surveillance in the studies included in the Cochrane Database.
In one study, the definition of high intensity was starting surveillance at six months versus one year for low intensity, while in another it was three months versus six months. Another study used only physical exams and defined high and low intensity as an ultrasound versus none, and in another, the definitions were CEA and CT scan versus none. Two other studies used definitions of CT scan and CEA versus CEA, and CT scan versus physical exams and CEA.
“It's all over the map,” Dr. Ryan said. “These studies are kind of thought pieces; they're not good data. They're nothing we would consider an argument for high-intensity screening.”
The Harm of Intensity
Radiation is also a huge issue when it comes to high intensity surveillance, Dr. Ryan said. “Every single patient I've seen with Stage IV colorectal cancer has asked me about the effects of radiation for CT scan.”
Renal damage also has to be considered, and if physicians do an abdominal CT scan should they also do a chest CT because lung metastases can be resected? And if so, how often should this scan be done—three months, six months, or annually?
Furthermore, are the issues of unwanted biopsies, the fact that there are rarely “truly negative” CT scans, and the anxiety that patients experience throughout the whole trajectory.
“We need a randomized controlled trial of CT scanning versus no CT scanning, or at least scanning at six months versus 12 months.”
Cost is a considerable factor when debating high-intensity vs low-intensity surveillance, Dr. Ryan said, demonstrating the cost implications for every 100 colorectal cancer patients:
For every 100 patients with Stage III colon cancer, 70 are cured and are receiving CT scans for no reason, he said. Of the remaining 30, 10 will likely present with liver-only disease and 20 with multisite disease. Of those 20 patients with multisite disease, assuming an effect size of about 30%, six of those patients may really only have liver metastasis, and of those it is likely that only two will be cured.
In dollars, 500 CT scans multiplied by $500 is $250,000 for every 100 patients. Multiplying that by the 35,000 Stage III colon cancer patients in the United States, the cost is $87.5 million—a large sum of money for a very small cure rate, Dr. Ryan noted. In Stage II disease, for every 100 patients, 85 are cured and 15 are not, and the actual number of patients is 60,000—nearly doubling the $87 million spent in Stage III disease.
The ultimate question when considering the harm of high-intensity surveillance, Dr. Ryan said: “Are we creating more illness with CTs than we are actually curing?”