November 2008: Volume 34, Number 11
Stool DNA Better Than Plasma DNA in Testing for Colorectal Cancer Mutations
The use of stool samples to test for mutated DNA revealed more genetic mutations than did the use of plasma, according to findings from a study of 25 patients who had undergone surgery for colorectal cancer (Gastroenterol 2008;135:489–98). Investigators led by Bert Vogelstein, MD, director of the Ludwig Center at Johns Hopkins University, used a “digital” assay, which quantifies template molecules individually with each positive reaction, to analyze stool specimens collected during surgery from participants in a previous study. The assay technology uses a process called BEAMing, which stands for beads, emulsions, amplification, and magnetic. Samples were assessed for 22 mutations in four target genes: APC, TP53, and K-ras, using a single base extension assay, and a sequencing approach for exons 9 and 20 of PIK3CA—as well as for exon 3 of CTNNBI, and exon 15 of APC. DNA was then sequenced in four separate reactions and enriched for the four target genes, whereafter the copy number of gene fragments was quantified. Plasma DNA was isolated and quantified via real-time PCR. The researchers then used the BEAMing assay for mutation analysis to test for 33 different mutations. They noted that for each stool specimen, 30,000 genome equivalents were analyzed. Ultimately, 92% of stool samples (23 of 25 samples) were found to have mutations present in the corresponding tumors from the same patient. However, DNA from plasma allowed detection of only 50% (8 of 16 samples). These results suggest that a noninvasive test for colon cancer risk could someday be considered as an adjunct to colonoscopy, the researchers wrote. However, more and larger studies will be needed, along with cost-effected methods for using the BEAMing assays to analyze a panel of target genes.
Continuous Glucose Monitoring Can Benefit Type 1 Diabetes Patients
Continuous glucose monitoring (CGM) in a randomized, multicenter trial of 322 adults and children age 8 years or older with type 1 diabetes was associated with improved glycemic control, researchers from the Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group reported (NEJM 2008; 359:1464–76). Patients were stratified by age—8–14 years, 15–24 years, and 25 years and older—and were assigned to CGM or home-based self monitoring of blood glucose (SMBG). Baseline HbA1c level for all patients was 7%–10%. At 26 weeks, CGM in all three age groups was associated with a larger reduction in HbA1c level, but it reached statistical significance only in the oldest cohort. For that group, 26% (13 of 52 patients) in the CGM arm had a relative decrease of at least 10%, versus only 4% (2 patients) in the SMBG control group. Likewise, for the adolescent/young adult group, a 10% or greater reduction was achieved in 14% of those assigned to CGM, compared with a 10% decrease in those on SMBG. Among patients age 8 years and younger, 29% on CGM had a 10% or greater drop in HbA1c, versus 12% of the children randomized to the control intervention. This trend was also true of the overall change over the entire 26-week period, but the only cohort for which there was a statistically significant decrease in glycated hemoglobin was the group age 25 years and older. These results indicate that CGM can help achieve a real benefit in HbA1c levels in adults, although more study is needed to determine how younger age groups can use this technology to reach the same benefit, the investigators concluded.
Mortality Rate Higher in Patients With High Troponin T but Without Chest Pain
Hospitalized patients with elevated levels of troponin T without chest pain had a greater mortality rate than patients with chest pain, according to a retrospective study from one emergency department. The investigators enrolled 92 patients who had a TnT level of > 0.04 µg/L, normal CK or CK-MB fraction < 5%, no evidence of MI or ischemia on electrocardiography, and no chest pain (Am J Cardiol 2008;102:668–71). The mean age of the cohort was 64 years. A control group of 91 patients with ACS, including chest pain, was also enrolled; the mean age for the control patients was 61 years. Exclusion criteria were end-stage kidney disease, left ventricular ejection fraction <40%, age younger than 18 years, or any critical illness. At 30 days, the study group had a significantly greater mortality rate of 13.0%, versus 4.4% for those in the control group. The final diagnoses in the study group comprised a number of serious conditions. Infection was diagnosed in 24%, followed by exacerbation of COPD, hypertensive urgency, GI bleed, hyperglycemia or ketoacidosis, dehydration, cerebral-vascular accident, and others. These results suggest that TnT may be a marker for increased risk of non-cardiac-related deaths, the investigators concluded.
Persistent Elevation of cTn I in Patients after ACS Episode
Levels of NT-proBNP measured 6 months after an ACS episode were the best predictor of persistently high levels of cTnI, according to findings from a recent randomized, controlled trial (Am Heart J 2008;156:588–94). Researchers enrolled 898 patients form the FRagmin and Fast Revascularization during Instability in Coronary Artery Disease (FRISC) II Trial. Each patient was assessed at 6 weeks, 3 months, and 6 months after randomization and followed for at least 5 years. A total of 233 patients showed persistent cTnI elevation, defined as more than 0.01 µg/L at all three time points; 377 were negative for cTnI, and 288 patients had a temporary elevation. Persistent elevation was independently predicted by levels of NT-proBNP at 6 months, male gender, and randomization to an early invasive strategy. NT-proBNP levels at 6 months were a mean of 171 ng/mL in cTnI-negative patients, 238 ng/mL in patients with a temporary elevation, and 488 ng/mL in those with a persistent cTnI elevation. There also was a significant trend toward lower left ventricular ejection fraction in patients with persistently elevated cTnI. The researchers concluded that impaired left ventricular function is the principal contributor to cTnI elevation in this setting.