Validated Colorectal Cancer Risk Assessment Tool Available Online

NCI researchers have developed one of the first colorectal cancer (CRC) absolute risk assessment models for people aged 50 to 85 without known genetic susceptibility to the disease, using data from two large case-control studies and validating the model in the NIH-AARP Diet and Health Study population. The tool uses answers about lifestyle, screening, and family history of CRC to calculate a person’s 5-year, 10-year, and lifetime risk of developing the disease. Both the development and validation papers were published December 29 in the Journal of Clinical Oncology, and an online risk assessment tool based on the model is now available for use by physicians at http://www.cancer.gov/ColorectalCancerRisk/.

“This colorectal cancer risk model provides physicians and their patients with a new tool to help make informed decisions about colorectal cancer screening and other prevention strategies. It may also assist policy makers in evaluating the usefulness of current and future population colorectal cancer prevention approaches,” said Dr. Andrew Freedman, lead author of the study, in an accompanying press release.

Dr. Freedman of NCI’s Division of Cancer Control and Population Sciences and Dr. Ruth Pfeiffer of NCI’s Division of Cancer Epidemiology and Genetics (DCEG) led the team that developed the model using data from two population-based, case-control studies, one for colon cancer and one for rectal cancer, and cancer incidence rates from the SEER database. Since non-Hispanic white men and women aged 50 years or older made up the majority of participants in both case-control studies, the researchers restricted their model to this group.

They determined the relative risks for a cancer diagnosis in three different areas of the colon and rectum based on known risk factors, including history of CRC screening, body mass index, and family history of CRC. The researchers also constructed a short risk-assessment patient questionnaire that easily captures the information used in the model.

To assess the model’s performance, the researchers, led by DCEG’s Dr. Yikyung Park, used the model to estimate the number of CRC cases expected in the 155,345 men and 108,057 women in the NIH-AARP cohort. They then compared those results to the actual number of cases diagnosed in that cohort between 1995 and 2003. They found that their model was well calibrated, with the estimated-to-observed ratio for CRC diagnoses being 0.99 in men and 1.05 in women.

The researchers are in the process of updating the Web tool with data from SEER to improve accuracy for other racial and ethnic groups.

Colonoscopy Reduces Death from Cancer in the Left Colon, but Not the Right

In a large case-control study performed by researchers at the University of Toronto, patients who had undergone colonoscopy at least once were significantly less likely to die of colorectal cancer (CRC) arising in the left side of the colon (the part of the colon closest to the rectum) than patients who had never had a colonoscopy. However, colonoscopy did not reduce mortality from CRC in the right side of the colon, according to results published in the December 15 Annals of Internal Medicine.

Using four Ontario-based databases, the investigators matched 10,292 patients who had died of CRC with 51,460 controls who did not have CRC. Each patient was matched to five controls by age, sex, income, and location of residence, and all controls were alive at the date of the patient’s death.

Because the investigators could not distinguish screening colonoscopies from diagnostic colonoscopies using the database records, they excluded all colonoscopies done within 6 months before a diagnosis of CRC.

Seven percent of case patients and almost 10 percent of controls had undergone colonoscopy. “Case patients were less likely than controls to have undergone any attempted colonoscopy…or complete colonoscopy,” stated the authors. Both attempted and complete colonoscopy reduced the likelihood of death from left-sided CRC, but not right-sided CRC.

The reasons for the lack of efficacy in preventing right-sided CRC are not clear, but may include less effective bowel preparation or colonoscopy technique in the right colon or differences in the biology and natural history of cancer arising in the right colon, explained the authors.

These results “seem consistent with a CRC mortality risk reduction of about 60 percent to 70 percent for the left colon and highlight the fact that we have very little data…about efficacy in the right colon,” said Dr. David Ransohoff of the University of North Carolina at Chapel Hill, in an accompanying editorial.

PLCO Update Suggests Positive Predictive Value of Prostate Screening Drops over Time

In the most recent update of prostate cancer screening data from NCI’s PLCO trial, the number of positive screening tests among the more than 38,000 men enrolled in the intervention arm has remained consistent during the 4 years of study. However, the positive predictive value of these tests—that is, the ratio of true positives to the total number of true and false positives—decreased over time. The full report appeared in the December issue of the British Journal of Urology International.

Researchers attributed this decrease in the positive predictive value from baseline screening, which was 17.9 percent, to the subsequent annual screens, which ranged from 10.4 percent to 12.3 percent, to the fact that initial screening identified many of the cancers and left men with elevated prostate-specific antigen (PSA) levels but negative biopsies to be identified in subsequent screening rounds. The cancers found at baseline were more likely than those found in subsequent screenings to be advanced stage (5.8 versus 2.9 percent) and to have a Gleason score of 7 to 10 (34 versus 25.5 percent).

PLCO is one of two major ongoing trials expected to show whether screening for prostate cancer saves lives. The other is the European Randomized Study of Screening for Prostate Cancer (ERSPC), which screens participants less frequently, at 2- and 4-year intervals. In a related commentary, ERSPC director Dr. Fritz H. Schröder addressed the issue of more aggressive prostate cancer, noting that the data in both trials “suggest that certain cancers escape screening irrespective of the screening interval used.”

Radiation Plus Hormone Therapy for Locally Advanced Prostate Cancer Improves Survival

A randomized clinical trial, published online December 15 in The Lancet, that tested hormone therapy (HT) alone versus HT plus radiation therapy (RT) for locally advanced prostate cancer adds to the body of evidence that adding RT to HT is more effective than HT alone for this group of patients.

Investigators from two European research groups, led by Dr. Anders Widmark of Umeå University in Sweden, assigned 875 men with locally advanced prostate cancer—cancer that had invaded local tissue but had not spread to the lymph nodes or distant sites—to receive either HT alone or HT plus standard 3D conformal RT. HT consisted of leuprorelin (known in the United States as leuprolide) given for 3 months and flutamide given until disease progression or death. Patients could switch to the drug bicalutamide if they experienced unacceptable side effects while taking flutamide.

After a median follow up of 7.6 years, 18 percent of the men in the HT group had died of prostate cancer compared to 8.5 percent of the men in the HT-plus-RT group. Ten-year mortality from prostate cancer was 23.9 percent for men in the HT group versus 11.9 percent in the HT-plus-RT group.

Follow-up visits revealed “a small but significant increase of moderate to severe late effects related to urinary and sexual function” in the HT-plus-RT group, stated the authors. However, patients did not report significant differences in their overall health and quality of life on surveys taken 4 years after treatment, with the exception of social function, which was decreased in the HT-plus-RT group.

Researchers Learn Why Some Platinum Drugs Are Toxic to Ear Tissue

Ototoxicity, or damage to inner-ear cells that detect sound, is a known side effect of platinum-based chemotherapy drugs. The damage is much more common with cisplatin than with oxaliplatin, though the reason for this difference has been unknown. Swedish researchers have now tested some of the possible explanations by comparing the activity of these two drugs in the inner ears of guinea pigs. Their analysis appeared December 30 in the Journal of the National Cancer Institute.

Cisplatin and oxaliplatin are thought to work by attaching to DNA and binding the genetic material so that it cannot replicate properly, thereby preventing cell division. But because the cells in the inner ear are already differentiated and less susceptible to this mechanism, the researchers hypothesized that cisplatin and oxaliplatin may also have an effect outside of the nucleus by generating reactive oxygen species, or free radicals.

They cultured colon carcinoma cells and measured the extent to which each drug induced cell death, or apoptosis, and found that cisplatin is significantly more dependent on free radicals than is oxaliplatin for triggering death in these cells. Guinea pigs showed more signs of free radical-induced apoptosis in their cochlea after exposure to cisplatin than to oxaliplatin.

The two drugs also showed marked effects on the transmission of sound signals from the cochlea to the brain, with cisplatin causing more damage to the hair cells that detect vibrations, and ultimately causing deafness, while comparable exposure to oxaliplatin did not. Cisplatin absorbed into the inner ear more easily than oxaliplatin did, and it remained in the tissues for a longer time.

“[It] should not be taken for granted that human subjects have the same cochlear pharmacokinetics [as guinea pigs],” the researchers noted, though they believe that the major aspects of their conclusions are valid for humans.