Over the past 2 decades, we have seen a substantial increase in the 5-year survival of patients with stage II and III colon cancer, marking an evolving oncologic success story. However, in the postoperative setting, the value of regular CT screening to monitor for recurrence has been greeted with mixed opinions.1 The ASCO Post recently asked two nationally regarded gastrointestinal cancer experts if CT scanning is appropriate in the routine follow-up of patients with colon cancer.
PRO: Richard M. Goldberg, MD
Physician-in-Chief and Professor
The Klotz Family Chair in Cancer Research
The Ohio State University Comprehensive Cancer Center
James Cancer Hospital and Solove Research Institute
Colorectal cancer is fairly unique among adult solid tumors in that recurrence generally occurs in the liver and lung or abdominal cavity, anatomic areas where metastases can be identified by scanning long before they would cause symptoms. Moreover, while we have a colon tumor marker, carcinoembryonic antigen (CEA), it is only elevated in about 60% of patients with recurrent disease and may be increased in some who never manifest recurrence, so it is a less-than-perfect method to detect recurrence. And the natural history of colorectal cancer is such that—at least in a subset of patients with recurrent disease—there are a limited number of metastatic deposits limited to a single organ or site.
Improved Surgical Outcomes
This characterization might seem counterintuitive because studies show circulating tumor cells in the bloodstreams of these patients, but very few of these cells actually establish metastatic deposits. So resecting the few liver metastases in patients with so-called “oligometastatic disease” can result in cure of metastatic disease.
Liver surgery has vastly improved over the past couple of decades, in part due to better technology but also due to the training of liver surgeons as part of a surgical oncology fellowship, leading to better and better outcomes. In fact, resecting limited liver metastases in selected patients with recurrent colorectal cancer can result in a 30% to 60% probability of long-term disease-free survival, even when the initial surgery to resect the primary tumor and drug therapy prove not to be curative. In a smaller subpopulation, surgical specialists can also resect a limited number of lung metastases, although this clinical scenario is much less common.
The liver is a large, thick organ in which physical examination through the abdominal wall muscles and liver function tests are most often useless in detection of resectable lesions. The point of performing regular CT surveillance is to detect liver metastases, many of which can be successfully resected.
On the other side of the equation, there are a few potential downsides to CT screening. A very small percentage of patients will have an anaphylactic reaction to the CT scan dye. That said, it’s a situation that scanning centers are adept at managing. There is also the chance that radiation exposure from CT can result in a future cancer, although large studies of Medicare data suggest that the risk is hard to quantify, and probably more theoretical than actual.2,3
The other concern, of course, is the cost of regular-interval screening, which is a value judgment that individual oncologists, their patients, and payers must make. The Cochrane Collaboration did a study on follow-up CT screening in stages II/III colon cancers, finding that costs per life saved were favorable according to the standard benchmarks for health-care expenditures.3
It is important to note that over the same period that we’ve seen treatment advances, the cost of CT screening has dropped considerably, and the procedure has become faster and more streamlined, reducing the amount of radiation exposure to the patient to a much safer level.
The decrease in deaths from recurrent colorectal cancer that we’ve seen in the management of metastatic disease is due mainly to surgical resection of limited recurrence, which can make the difference between death or cure. We’ve also seen considerable progress in the use of cytotoxic drugs and targeted therapies. I believe data provided by the Cochrane Collaboration, along with consensus-based clinical guidelines from ASCO and the National Comprehensive Cancer Network, support my opinion that CT scan surveillance for patients who have been treated for stage II or III colon cancer is sound practice in which the benefits outweigh the risks and costs.
Disclosure: Dr. Goldberg has received research funding from Pfizer.
CON: David P. Ryan, MD
Department of Medicine
Harvard Medical School
MGH Cancer Center
Massachusetts General Hospital
Everything we do in medicine should meet one of three goals: curing patients, extending life, or helping patients feel better. If an intervention is not accomplishing one of those core elements of care, then we shouldn’t be doing it. Although generally not considered an intervention, imaging is in fact an intervention that can cause discomfort—both physical and emotional—and side effects such as contrast-induced allergic reactions and nephropathy. In addition, from the mammography literature it is apparent that imaging leads to a significant number of false-positives, which lead to unnecessary biopsies and follow-up scans.
Unlike mammography, we do not have sufficient data to quantify harms associated with follow-up surveillance in the postoperative setting or the clinical value of regular-interval CT scans in stage II/III colon cancers to check for recurrence. So if we do not know for certain that scans are helping to cure these patients, it begs the question: What is our goal in this setting? Surveillance scans are not going to help people live longer or feel better.
The theory behind surveillance CT scans is to detect recurrence when the disease is still small and isolated to the liver so it is still resectable, which, according to survival data, will give the patient about a 30% chance of cure. However, despite the theoretical upside to this potentially curative scenario, there is no level 1 evidence demonstrating that we are actually curing people because of surveillance scanning. All we have are several small, poorly executed studies, a few of which suggest that CT scanning may detect metastatic disease sooner than in a no-screening scenario, but there is no evidence that it translates into higher cure rates.
Small Potential Benefit
If we were to design a randomized study assessing surveillance scans in the postoperative setting, how much could we reasonably expect the experimental arm of surveillance to improve survival? For every 100 patients with stage III colon cancer, 70 will be cured and 30 will develop recurrent disease. From the beginning, we know that surveillance scans will add harm in those 70 patients. For the 30 patients who will experience a recurrence, we expect that 20%, or 6 patients, will present with disease isolated to the liver—regardless of how they are followed postoperatively—and be offered a chance at curative resection.
Therefore, we are monitoring 24 patients out of every 100 in order to detect their disease at an earlier time point, when it might be isolated to the liver or lungs and be amenable to resection. If we estimate that surveillance CTs detect 30% more patients with isolated disease amenable to resection, this will impact only 7 of 100 patients, of whom only 2 to 3 will eventually be cured with surgical resection. If we aggregate the added MRIs, PET scans, and procedures, we can assume that the practice of surveillance CT scans costs millions of dollars for every life that is theoretically saved.
Given this extremely small potential benefit, it is incumbent upon the research community to prove to patients and providers that routine surveillance screening of patients after surgery for stage II/III colon cancer is worthwhile. ■
Disclosure: Dr. Ryan has served as a consultant or advisor to Threshold Pharmaceuticals, Array Bioscience, and Millennium Pharmaceuticals.
1. Goldberg RM, Ryan DP: Scan? Cure? Sure! Oncologist 16:254-256, 2011.
2. Meer AB, Basu PA, Baker LC, et al: Exposure to ionizing radiation and estimate of secondary cancer in the era of high-speed CT scanning: projections from the Medicare population. J Am Coll Radiol 4:245-250, 2012.
3. Jeffery M, Hickey BE, Hider PN: Follow-up strategies for patients treated for non-metastatic colorectal cancer. Cochrane Database Syst Rev 1:CD002200, 2007.
Interviews conducted and report compiled by Ronald Piana.