CA-125: To Monitor or Not to Monitor?
Dr. Rustin: I agree with much of what Dr. Karlan says about the possible limitations of the study, but I must take issue with a few points.
One criticism of the trial has been that patients were not stratified according to the extent of initial surgery. This was not done because patients entered the trial after completing first-line therapy. At the time of initial surgery, they were not in a trial and it was considered that the reporting of surgical results would have been difficult to interpret. Furthermore, what was then described as optimal surgery would, in many cases, now be considered suboptimal. However, there are 2 reasons why criticism regarding lack of information on initial surgery is misplaced. First, all patients had to be in complete remission with a normal CA-125 level to enter the trial. Thus, these patients were already in a better prognostic group, reducing the prognostic impact of the earlier surgery. Second, because 529 patients were randomly assigned between early or delayed chemotherapy for relapse, it is highly unlikely that there could have been a major imbalance in initial surgery between the 2 arms.
Dr. Markman: Is it possible that a more aggressive approach to secondary surgical cytoreduction in the presence of a rising CA-125 might have changed the study results, at least for a subset of patients?
Dr. Rustin: Cytoreductive surgery was performed on 21 patients in the early treatment arm and on 14 patients in the delayed treatment arm of the MRC/EORTC trial.The role of cytoreductive surgery for relapsed ovarian cancer remains uncertain. Only when we have the results of the randomized DESKTOP 3 and GOG 213 trials in a few years' time will we know whether there is any survival advantage from this surgery. However, already we are able to say that there are large numbers of patients who are very unlikely to benefit from surgery and who therefore will not benefit from routine CA-125 measurements during follow-up. These include all patients during the first 12 months of follow-up, because most gynecologic oncologists would not recommend cytoreductive surgery to patients who relapse within 12 months of their initial treatment. Patients who had suboptimal initial surgery or ascites or a performance score of > 0 were shown in the DESKTOP 1 and -2 trials to rarely achieve complete surgical clearance and therefore would not benefit from routine CA-125 measurements.
I am very concerned about doing routine CA-125 levels on all patients to detect surgically resectable disease. The great majority with a rising CA-125 level will not benefit from surgery but are made aware of recurrent cancer earlier than necessary, causing anxiety and a great chance of receiving chemotherapy earlier than necessary. Even in Dr. Karlan's unit only about 5 patients per year received secondary cytoreductive surgery. Presumably far more patients were told that their CA-125 level was rising and started chemotherapy months earlier than we now know was necessary. I am also concerned at the way retrospective data are interpreted. In the Fleming study, patients who underwent optimal secondary cytoreductive surgery went to the operating room 5.3 weeks from the time of their CA-125 elevation, and those with suboptimal surgery waited 16.4 weeks before surgery. Why did one group wait so long for surgery? Could it be that they delayed because initial scans did not show clear-cut operable disease?
Many of us have had patients who we are convinced benefited from secondary cytoreductive surgery, and until we have randomized trial data that confirm that it improves survival, we have to use whatever data are available as well as common sense to select patients for surgery. I can't see how a patient who had miliary peritoneal disease at diagnosis is likely to have just 1 or 2 isolated sites of disease at relapse. PET scans will not detect microscopic disease so they cannot be relied upon to exclude a considerable proportion of patients from unnecessary surgery, though PET will exclude those with multiple hot spots. I consider it highly unlikely that a higher rate of secondary cytoreductive surgery would have had any impact on survival in the MRC/EORTCtrial. Therefore, I continue to recommend against routine CA-125 measurements to detect surgically resectable disease but accept that a case could be made for performing a CA-125 measurement and CT scan in a selected group of patients who remain asymptomatic at least 1 year after completion of first-line therapy.
Dr. Markman: Beth, regarding the population of women who might be candidates for a secondary surgical cytoreduction -- where somewhat earlier recognition of recurrence would result in a legitimate attempt at removal of all gross intraperitoneal disease -- do you think this is a small percentage of patients, as suggested by Gordon, or do you suspect that this represents a much larger group of individuals?
Taking Issue With the Issue
Dr. Rustin: I agree with much of what Dr. Karlan says about the possible limitations of the study, but I must take issue with a few points.
One criticism of the trial has been that patients were not stratified according to the extent of initial surgery. This was not done because patients entered the trial after completing first-line therapy. At the time of initial surgery, they were not in a trial and it was considered that the reporting of surgical results would have been difficult to interpret. Furthermore, what was then described as optimal surgery would, in many cases, now be considered suboptimal. However, there are 2 reasons why criticism regarding lack of information on initial surgery is misplaced. First, all patients had to be in complete remission with a normal CA-125 level to enter the trial. Thus, these patients were already in a better prognostic group, reducing the prognostic impact of the earlier surgery. Second, because 529 patients were randomly assigned between early or delayed chemotherapy for relapse, it is highly unlikely that there could have been a major imbalance in initial surgery between the 2 arms.
Dr. Markman: Is it possible that a more aggressive approach to secondary surgical cytoreduction in the presence of a rising CA-125 might have changed the study results, at least for a subset of patients?
Dr. Rustin: Cytoreductive surgery was performed on 21 patients in the early treatment arm and on 14 patients in the delayed treatment arm of the MRC/EORTC trial.The role of cytoreductive surgery for relapsed ovarian cancer remains uncertain. Only when we have the results of the randomized DESKTOP 3 and GOG 213 trials in a few years' time will we know whether there is any survival advantage from this surgery. However, already we are able to say that there are large numbers of patients who are very unlikely to benefit from surgery and who therefore will not benefit from routine CA-125 measurements during follow-up. These include all patients during the first 12 months of follow-up, because most gynecologic oncologists would not recommend cytoreductive surgery to patients who relapse within 12 months of their initial treatment. Patients who had suboptimal initial surgery or ascites or a performance score of > 0 were shown in the DESKTOP 1 and -2 trials to rarely achieve complete surgical clearance and therefore would not benefit from routine CA-125 measurements.
I am very concerned about doing routine CA-125 levels on all patients to detect surgically resectable disease. The great majority with a rising CA-125 level will not benefit from surgery but are made aware of recurrent cancer earlier than necessary, causing anxiety and a great chance of receiving chemotherapy earlier than necessary. Even in Dr. Karlan's unit only about 5 patients per year received secondary cytoreductive surgery. Presumably far more patients were told that their CA-125 level was rising and started chemotherapy months earlier than we now know was necessary. I am also concerned at the way retrospective data are interpreted. In the Fleming study, patients who underwent optimal secondary cytoreductive surgery went to the operating room 5.3 weeks from the time of their CA-125 elevation, and those with suboptimal surgery waited 16.4 weeks before surgery. Why did one group wait so long for surgery? Could it be that they delayed because initial scans did not show clear-cut operable disease?
Many of us have had patients who we are convinced benefited from secondary cytoreductive surgery, and until we have randomized trial data that confirm that it improves survival, we have to use whatever data are available as well as common sense to select patients for surgery. I can't see how a patient who had miliary peritoneal disease at diagnosis is likely to have just 1 or 2 isolated sites of disease at relapse. PET scans will not detect microscopic disease so they cannot be relied upon to exclude a considerable proportion of patients from unnecessary surgery, though PET will exclude those with multiple hot spots. I consider it highly unlikely that a higher rate of secondary cytoreductive surgery would have had any impact on survival in the MRC/EORTCtrial. Therefore, I continue to recommend against routine CA-125 measurements to detect surgically resectable disease but accept that a case could be made for performing a CA-125 measurement and CT scan in a selected group of patients who remain asymptomatic at least 1 year after completion of first-line therapy.
Dr. Markman: Beth, regarding the population of women who might be candidates for a secondary surgical cytoreduction -- where somewhat earlier recognition of recurrence would result in a legitimate attempt at removal of all gross intraperitoneal disease -- do you think this is a small percentage of patients, as suggested by Gordon, or do you suspect that this represents a much larger group of individuals?
SHARE
![Breast Cancer Screening (PDQ®): Screening - Health Professional Information [NCI]-Special Population](https://cdnus.oss-us-west-1.aliyuncs.com/libs/images/NewsJpg/cate_6/039_215x160.jpg)
