The premise that early diagnosis of certain types of malignancies improves outcome and survival is a cornerstone of modern medicine. Routine use of the Pap smear has been associated with reduced mortality from cervical cancer. Randomized trials have demonstrated a survival benefit for screening mammography for women aged 50 to 70 years. Similarly, data have emerged to support screening for colorectal cancer.
Use of prostate-specific antigen (PSA) screening for prostate cancer has also been endorsed, although evidence supporting this practice has been derived from nonrandomized trials. All of these practices are predicated on the tenet that early diagnosis of certain cancers saves lives.
As a consequence of the success of screening practices, there is an understandable belief by both patients and physicians that early diagnosis of recurrent diseasewhen it is presumably more amenable to treatmentwill also yield a clinical benefit. Frequently, this translates into regular, intensive testing of asymptomatic survivors of early-stage cancers in the hope that early detection of metatastic disease will improve survival and/or quality of life.
Lack of Benefit From Screening
Unfortunately, this belief has not been borne out. Indeed, as Drs. Schwartz, Billingsley, and Wallner discuss, existing data show that rigorous testing of survivors of early-stage breast, prostate, or colorectal cancer has little impact on their ultimate outcome. Further, such testing is quite costly and places a substantial burden on the health-care system, especially since these three malignancies are diagnosed in over 400,000 Americans each year.
Evidence against routine screening for metastatic breast cancer has been most thoroughly established. Two large randomized trials that included more than 2,500 patients have demonstrated that more intensive follow-up may detect recurrent disease sooner. However, both studies failed to show that earlier detection translates into improved survival or a delay in cancer-related deaths.[1,2] Further, the GIVIO (Interdisciplinary Group for Cancer Care Evaluation [Gruppo Interdisciplinare Valutazione Interventi Oncologia]) trial showed that patient-assessed quality of life was similar in the intensive and nonintensive groups.
Importantly, 70% of recurrences were detected by patients between visits.[1,2] Proponents of regular, intensive screening point out that neither of these trials examined the use of newer tests, such as tumor markers or computed tomography. Further, these trials were conducted before certain newer salvage therapies, such as the taxanes, aromatase inhibitors, or trastuzumab(Drug information on trastuzumab) (Herceptin), became available.
However, it seems unlikely that these interventions would result in substantially different findings, particularly in an era where adjuvant systemic therapy is widely used. Given the results of these two trials and other data cited by Schwartz et al, recent evidence-based guidelines from the American Society of Clinical Oncology (ASCO) and an Italian consensus panel appropriately recommend only regular office visits, with testing limited to a targeted evaluation of symptoms or physical abnormalities that are suggestive of metastatic disease.[3,4]
The only exception is mammography. By extrapolation from screening studies, mammography to detect ipsilateral breast recurrence or a contralateral breast primary is the only suggested routine follow-up test in early breast cancer survivors, since treatment of these two entities is likely to be associated with improved long-term survival.
High Price of Follow-up Screening
The range and costs of follow-up for colorectal cancer vary even more widely than those for breast cancer. In addition, randomized data addressing the value of screening of colorectal cancer survivors are considerably more limited, involving about 500 patients enrolled in three trials.[5-7] However, these trials involve several diagnostic modalities, including computed tomography and colonoscopy.
These and other available data were reviewed by ASCO last year; their evidence-based guidelines center on office visits every 3 to 6 months for the first 3 years, and annually thereafter. Monitoring of carcinoembryonic antigen is warranted only in stage II or III patients in whom hepatic metastectomy would be considered. Surveillance for local recurrence by sigmoidoscopy is recommended for patients with rectal cancer who do not receive radiation therapy.
Even less information is available about the value of follow-up after treatment of early prostate cancer. No randomized evaluation of follow-up testing has been performed. This is somewhat surprising, since a rather sensitive and specific test existsPSA. Here, expert opinion appears to support PSA monitoring to identify the fraction of patients who develop local recurrence that might be amenable to salvage surgery or radiotherapy. Neither digital rectal examination nor routine imaging studies appear to offer value in asymptomatic individuals or in the absence of a rising PSA. Thus, PSA testing and office visits to elicit symptoms of disease recurrence or therapy complications should also be the focus of prostate cancer follow-up at this time.[9,10]
For some time, cancer specialists have realized that more is not always better. This appears to be the case for follow-up testing, since prospective trials have not demonstrated advantages for intensive or sophisticated testing of breast and colorectal survivors, and a paucity of information exists for the follow-up of prostate cancer survivors.
As the cost of health care continues to escalate, it is imperative that we as physicians employ evidence-based medicine, rather than intuition in our daily practice. In the case of follow-up testing, this means shifting our priorities to regular patient assessment and selective laboratory testing. Adherence to this lower-cost strategy does not appear to endanger patient outcome; rather, it should help optimize the allocation of health-care dollars and ensure that funds are available to develop and provide better strategies for cancer screening, prevention, and treatment.