Circling Back: PSA Screening for Colorectal Cancer Patients

Ayal A. Aizer, MD, MHS

Points Made in “Should All Colorectal Cancer Patients Over Age 60 Be Screened for Prostate Cancer?” Commentaries Are Addressed by Author Aizer

In the October 2013 issue of ONCOLOGY, Dr. Ayal Aizer of the Harvard Radiation Oncology Program and Dr. Anthony D’Amico of the Dana-Farber Cancer Institute publisheda paper that addressed the question of whether all male colorectal cancer patients over the age of 60 should be screened for prostate cancer. As with all ONCOLOGY articles, Dr. Aizer and Dr. D’Amico’s review was accompanied by two commentaries that presented different perspectives on this question. Because the authors of these commentaries made some especially interesting points, we wanted to provide Dr. Aizer and Dr. D’Amico with a chance to respond. We’re joined here today by Dr. Aizer, who will address some of these points.

-Interviewed by Susan Beck

ONCOLOGY: The authors of the first commentary, Drs. Brian Davis, Thomas Pisansky, and William Catalona, make the point that, when trying to determine whether PSA screening makes sense for a particular man with colorectal cancer, it’s important to consider not just the median life expectancy for his particular cohort, but also the statistical distribution around the mean expectancy. For example, even though the average life expectancy for a patient of a given age may be 8 years (which might exclude him from PSA screening according to your criteria), perhaps 15% of patients that age will actually live another 14 years. Would you agree that it’s important to consider the distribution as well as the mean life expectancy-and if so, how, in your view, would that affect the decision-making process?

Dr. Aizer: Thank you for the question. We appreciate the point that’s been made by Dr. Davis and colleagues, which really has allowed us to further clarify this issue. So, as mentioned, life expectancy estimates typically refer to a median. In actuality, there are patients who are going to exceed these expectations and others that will not, and it’s important to consider the standard deviation of the estimate and to counsel patients accordingly. I think this also brings up the point that we need better and more accurate estimates of life expectancy, and we should be using information such as age and comorbidity to make determinations related to life expectancy. The National Comprehensive Cancer Network (NCCN) actually has an approach where a patient’s age-expected life expectancy is adjusted based on their comorbidity level. Men in the top quartile of comorbidity typically display 50% improved survival, based on what is predicted by their age, and men in the bottom quartile of comorbidity, meaning those who are the most sick, display a 50% decrement. For example, if a man is 76 years of age and has a 10-year life expectancy based on age estimates alone, but they’re in the top quartile of health, then we would anticipate that their survival would be closer to 15 years. However, as mentioned by Dr. Davis and colleagues, calculating the distribution around this estimate is very important. So we really appreciate the opportunity to clarify this point, and we agree that it’s not only the median or mean estimate for life expectancy but the distribution around it that doctors and patients should be considering when determining whether to screen for prostate cancer using PSA.

ONCOLOGY: Thanks a lot. Dr. Davis and colleagues also make the point that patients with stage IV colorectal cancer shouldn’t be automatically excluded from PSA screening, since some of these patients-if they have limited hepatic metastases and undergo metastasectomy and systemic therapy-will have significant 5- and 10-year survival rates. What are your thoughts on PSA screening for this particular patient population?

Dr. Aizer: This is another great point made by Dr. Davis and colleagues-and they are correct: a significant percentage of men with colorectal cancer and isolated resectable liver metastases will be long-term survivors, and the current data suggest that approximately 40% of such patients may be free of any colorectal cancer at 5 years. Many of these men will ultimately be cured, and therefore, when determining the appropriateness of PSA screening for prostate cancer, we would favor including such men, meaning men with isolated, resectable liver metastases, but no other signs of other metastases, with the men who are in the stage III colorectal cancer cohort, meaning that men with stage III colorectal cancer may be similar to those with isolated resectable liver metastases, and therefore, if they achieve a disease-free interval of 5 years, PSA screening should be considered in such men because they very well may be cured of their colorectal cancer.

ONCOLOGY: Thank you. Another point Dr. Davis and colleagues make-actually it’s several-they enumerate a couple of purposes for PSA screening other than simply extending overall survival. These include:

• First, among patients who have never before undergone PSA testing, identifying those with extensive asymptomatic prostate cancer, so that they can receive androgen deprivation therapy and thus reap the significant quality-of-life benefits-and improved short-term life expectancy-that this therapy can provide.

• Two other purposes they mentioned were both to rule out prostate cancer as a coincident diagnosis and to identify those patients who actually do have concomitant prostate and rectal cancers-so that in both these cases, the most appropriate management can be provided.

What do you think of these additional purposes for PSA screening in patients with colorectal cancer?

Dr. Aizer: We agree that Dr. Davis and colleagues have made another outstanding observation here. When we wrote our article, we used survival as the metric to guide screening. We did not really base our recommendations on quality of life, although we agree that this is an important endpoint. And this is actually something that’s been espoused by other authors as well. Dr. Pamela Hartzband and Dr. Jerome Groopman wrote in a New England Journal of Medicine piece entitled “There Is More to Life Than Death” that bone pain, pathologic fractures, and urinary obstruction all are important measures for patients with prostate cancer, and one can envision that one benefit of PSA screening may be to identify men with advanced prostate cancer who are currently asymptomatic, so that they can receive therapy such as androgen deprivation therapy, and minimize the risk of skeletal events, pain, and urinary dysfunction. It’s also true that another benefit for screening for prostate cancer, in patients with newly diagnosed rectal cancer, for example, is that the treatment of rectal cancer may compromise the ability to treat prostate cancer in the future. For example, if patients receive radiation in the process of their rectal cancer treatment, it may be difficult to proceed with radiation in the treatment of a later-diagnosed prostate cancer. So this provides more of an impetus to screen for prostate cancer at the time of diagnosis of rectal cancer, so that the two cancers can be treated in a synchronized and coordinated manner. So ultimately what we would say is, if a man with newly diagnosed colorectal cancer is over 60 years old and has never had a PSA, we would screen stage I and II colorectal cancer patients for prostate cancer at diagnosis, especially if their life expectancy exceeds 10 years, after accounting for age and comorbidity. If there’s a patient who is diagnosed with colorectal cancer that is stage III or limited stage IV, meaning isolated and resectable hepatic metastases, then the patient should be aware that if they are screened for prostate cancer and are diagnosed with either a low-risk or a favorable intermediate-risk prostate cancer, that they probably should be undergoing active surveillance for that condition. However, if that prostate cancer is high-intermediate–risk or high-risk, then it’s more likely that it would be appropriate to treat both their prostate cancer and their rectal cancer simultaneously.

ONCOLOGY: Let’s move on to the second commentary. The authors of that commentary, Drs. Jeffrey Olsen and Jeff Michalski, raise the point that there are unique challenges involved in diagnosing and treating prostate cancer in patients with colorectal cancer. One of these challenges, they contend, is that PSA levels may be lower if a patient has received radiation therapy for rectal cancer. So how would you propose approaching PSA screening in a man who has rectal cancer and has already received radiation therapy by the time you see him?

Dr. Aizer: This is a great point, and what we would say is that patients who received radiation for rectal cancer may still be candidates for PSA screening for prostate cancer. What we would favor, however, is that rather than use the absolute number for the PSA to guide whether a biopsy should be pursued or not, we would recommend using the rate of change, and this is something that has been espoused by the NCCN as well. So the rate of change we would not expect to be altered by the prior pelvic radiation, and if that rate of change was more than 0.35 ng/mL per year, as stated by the NCCN, then this should be worrisome to practitioners and should warrant consideration of a prostate biopsy. So ultimately, it’s the rate of change that may be most appropriate for such patients.

ONCOLOGY: That’s interesting. Olsen and Michalski also note that in patients with rectal cancer who have undergone perineal resection, digital rectal examination and transrectal ultrasound-guided biopsy are not possible. How would you approach the question of PSA screening in this patient population, given the difficulties in pursuing normal follow-up should there be a suspicious result?

Dr. Aizer: This is another great question. What we would say is that it’s still important to pursue PSA-based screening for prostate cancer, but when it comes to the biopsy, what we would favor is an MRI-guided transperineal biopsy, because that would be the most feasible way to get tissue from the prostate. And now, in the era of template-guided biopsies that can be performed transperineally, this could be a very viable option for making a diagnosis of prostate cancer in patients who have a suspicious PSA.

ONCOLOGY: And the last point that Olsen and Michalski make is that surgical and radiation therapy options for treating early-stage prostate cancer may be limited if a patient has already had pelvic surgery or radiation therapy to the pelvis to treat his colorectal cancer. How would you approach the PSA screening question in these patients?

Dr. Aizer: We appreciate this question as well. I think Dr. Olsen and Dr. Michalski are very much correct. What we would say is that the treatment of rectal cancer may compromise the ability to treat subsequent prostate cancer. For example, the prostate may have received radiation in the process of rectal cancer treatment, and it may be difficult to readminister radiation to the prostate if they develop a prostate cancer. In addition, patients who have had an abdominoperineal resection (APR), meaning a rectal cancer surgery, may be less desirable candidates for prostatectomy, given the scarring that can be associated with that procedure. So what we would say is that in such scenarios, meaning patients treated for rectal cancer now with prostate cancer, it may be more appropriate to consider techniques such as external beam radiation for their prostate cancer, or either high-dose-rate (HDR) or low-dose-rate (LDR) brachytherapy-and if brachytherapy is going to be considered, then it would be important to use MRI guidance to facilitate the brachytherapy. There are some centers in the United States that have intraoperative MRI suites where this could be readily accomplished.

ONCOLOGY: Thank you very much!