Citations

Abstract Carfilzomib is a selective proteasome inhibitor that binds irreversibly to its target. In phase 1 studies, carfilzomib elicited promising responses and an acceptable toxicity profile in patients with relapsed and/or refractory multiple myeloma (R/R MM). In the present phase 2, multicenter, open-label study, 129 bortezomib-naive patients with R/R MM (median of 2 prior therapies) were separated into Cohort 1, scheduled to receive intravenous carfilzomib 20 mg/m2 for all treatment cycles, and Cohort 2, scheduled to receive 20 mg/m2 for cycle 1 and then 27 mg/m2 for all subsequent cycles. The primary end point was an overall response rate (≥ partial response) of 42.4% in Cohort 1 and 52.2% in Cohort 2. The clinical benefit response (overall response rate + minimal response) was 59.3% and 64.2% in Cohorts 1 and 2, respectively. Median duration of response was 13.1 months and not reached, and median time to progression was 8.3 months and not reached, respectively. The most common treatment-emergent adverse events were fatigue (62.0%) and nausea (48.8%). Single-agent carfilzomib elicited a low incidence of peripheral neuropathy—17.1% overall (1 grade 3; no grade 4)—in these pretreated bortezomib-naive patients. The results of the present study support the use of carfilzomib in R/R MM patients. This trial is registered at www.clinicaltrials.gov as NCT00530816.

Background— Whether mitral valve repair during coronary artery bypass grafting (CABG) improves survival in patients with ischemic mitral regurgitation (MR) remains unknown. Methods and Results— Patients with ejection fraction ≤35% and coronary artery disease amenable to CABG were randomized at 99 sites worldwide to medical therapy with or without CABG. The decision to treat the mitral valve during CABG was left to the surgeon. The primary end point was mortality. Of 1212 randomized patients, 435 (36%) had none/trace MR, 554 (46%) had mild MR, 181 (15%) had moderate MR, and 39 (3%) had severe MR. In the medical arm, 70 deaths (32%) occurred in patients with none/trace MR, 114 (44%) in those with mild MR, and 58 (50%) in those with moderate to severe MR. In patients with moderate to severe MR, there were 29 deaths (53%) among 55 patients randomized to CABG who did not receive mitral surgery (hazard ratio versus medical therapy, 1.20; 95% confidence interval, 0.77–1.87) and 21 deaths (43%) among 49 patients who received mitral surgery (hazard ratio versus medical therapy, 0.62; 95% confidence interval, 0.35–1.08). After adjustment for baseline prognostic variables, the hazard ratio for CABG with mitral surgery versus CABG alone was 0.41 (95% confidence interval, 0.22–0.77; P =0.006). Conclusion— Although these observational data suggest that adding mitral valve repair to CABG in patients with left ventricular dysfunction and moderate to severe MR may improve survival compared with CABG alone or medical therapy alone, a prospective randomized trial is necessary to confirm the validity of these observations. Clinical Trial Registration— URL: http://www.clinicaltrials.gov . Unique identifier: NCT00023595.

Abstract CN03-03 Multiple myeloma is the second most common hematological malignancy, affecting nearly 20,000 individuals each year with nearly 12,000 deaths during the same period. While considerable progress has been made in the treatment of this disease and the survival has improved in the recent years, myeloma still remains incurable. Given the incurable nature of the disease there has been considerable interest in understanding the pathogenesis and evolution of monoclonal gammopathies, which can improve the prospects for effective prevention strategies. Monoclonal gammopathy of undetermined significance has been recognized as precursor state for development of multiple myeloma for years. The prevalence of MGUS increases with age and forms nearly two thirds of the patients with a diagnosis of monoclonal gammopathy. The risk of progression from MGUS to myeloma has studied in several large cohort studies and is estimated to be around 1% per year. This risk remains fairly constant over the lifetime, highlighting the possibility that a ‘second hit’ is required in an individual with MGUS for its evolution to myeloma. Several clinical features have been identified that increases the risk of progression in any given individual, including the immunoglobulin isotype, M protein concentration, circulating plasma cells, and suppression of uninvolved immunoglobulin. More recently, the introduction of assays that can detect free light chains unbound to the heavy chain has allowed better identification of patients at increased risk of progression to myeloma. In fact, a staging system that incorporates the type of immunoglobulin, its concentration and presence of abnormal light chain ration allows us to identify a proportion of patients that are at high risk of progression. However, more recent epidemiologic studies suggest that nearly all patients with myeloma have a MGUS phase prior to the development of myeloma. This finding reinforces the importance of evaluating strategies that prevent the progression from MGUS to symptomatic myeloma. In addition, many patients are diagnosed with smoldering myeloma, a condition that falls between MGUS and symptomatic myeloma, and one which is typically observed with out any proven effective interventions. As in MGUS, there are several features that have been identified that allow risk stratification these patients. While patients with smoldering myeloma have several fold risk of progression compared to MGUS, it provides another point in the disease spectrum where prevention strategies may be valuable. The biggest obstacle to pursuing prevention strategies in patients with MGUS has been the low overall risk of progression, thus compromising the risk benefit ratio for most available anti-myeloma strategies to be used in a preventive fashion. Ongoing studies, especially that are examining potential genetic markers of progression, may allow better determination of an individual’s risk allowing us to focus on the small proportion of individuals with the highest level of risk. Another option would be to examine interventions with none or minimal adverse effects, that would allow a more universal approach to prevention. These could include vaccination strategies to modulate the immune system, use of food supplements and other measures which are being currently being evaluated. Given the much higher risk of progression among patients with smoldering myeloma, there have been clinical trials evaluating the role of anti-myeloma drugs as a strategy for preventing progression to symptomatic myeloma. Thalidomide has been studied with variable success. Clinical trials are being planned to evaluate the role of lenalidomide, an immunomodulatory analogue of thalidomide. Citation Information: Cancer Prev Res 2008;1(7 Suppl):CN03-03.

Abstract B39 Background The breast glandular epithelium is composed of lobules that undergo age-related involution that increases as women age and has been associated with lowered breast cancer risk. Mammographic density (MD) is an established risk factor for breast cancer that has been shown to decrease with age. This study examined the association of age-related lobular involution with mammographic breast density. Materials and Methods The study sample was selected from approximately 20,000 healthy women who were enrolled in a screening cohort at the Mayo Clinic (Mayo Mammography Health Study) from 2003 to 2006 and had a benign breast biopsy (BBD) within 2 years of the enrollment mammogram. A quantitative measure of mammographic density, percent density (PD) was estimated on the enrollment mammogram, in addition to its components dense and nondense area (NDA). All biopsies were reviewed by a breast pathologist for the type of benign breast disease (BBD) as well as lobular involution status. Involution was defined as the proportion of terminal ductal lobular units (TDLUs) in the background tissue on biopsy that showed at least 50% decrease in size and number of acini from normal lobules and was subjectively assessed as: No- 0% involution, Partial- 1 to 74%, or Complete- >75% involution. Mean density was estimated for the 3 categories of involution using generalized linear models and square root transformation. Confounders were assessed including age, parity, body mass index (BMI), type of BBD, menopausal status, age at first birth and family history Results A total of 317 women from the MMHS had BBD within 2 years of enrollment. Mean age at BBD diagnosis in this sample was 55 years. Age- and BMI- adjusted analyses suggested an inverse association of involution and percent density with adjusted mean PD values of 23.3 (95%CI [17.2, 30.3]) for no involution, 21.9 (95%CI [20.0, 23.9]) for partial involution, and 17.8 (95% CI [14.7, 21.3]) for complete involution (p for trend 0.05). This trend was driven primarily by the positive association of involution with nondense area with adjusted mean NDA values of 103.1cm2 (95% CI [86.1, 121.7]) for no involution, 104.9 cm2 (95% CI [99.4, 110.7]) for partial involution, and 126.2 cm2 (95% CI [115.0, 137.9]) for complete involution (p for trend 0.004). Discussion Our findings suggest that MD is inversely associated with involution of breast tissues, providing a plausible mechanism through which MD influences breast cancer risk. Citation Information: Cancer Prev Res 2008;1(7 Suppl):B39.

Abstract B24 Purpose Pancreatic cancer (PC) is the fourth leading cause of cancer death in the United States for which no recommended screening guidelines are available. An estimated 34,290 deaths related to PC is expected to occur in 2008 and the 5-year relative survival rate for this cancer is 5% for all stages combined. A person's chance of developing this cancer increases three-fold if a first-degree relative has pancreatic cancer. The purpose of this study was to assess attitudes of unaffected family members in the Mayo Clinic Pancreas Cancer Family study toward PC cancer risk and future screening options. This is the first study to report such attitudes of PC family members. Subjects and Methods At-risk family members with two or more relatives affected with pancreas cancer in the kindred and primary care controls completed a survey asking about: perceived PC risk, degree of PC worry/concern, attitude toward cancer screening in general, and intentions regarding PC screening if it were a blood test or endoscopic ultrasound. Results 386 family members in 109 PC kindreds and 1058 controls completed surveys. Forty-six percent of PC family members reported that it was likely they would develop PC during their lifetime compared to 6% of controls. Forty-one percent of family members reported having thoughts regarding the possibility of getting pancreatic cancer during the past month versus 4.6% of controls. Similarly, 74% of family members reported concern about getting PC as compared to 40.6% of controls. 94% of family members reported they would be likely to undergo a blood-based PC screening test compared to 89.7% of controls. For endoscopic ultrasound, 71% of family members would be likely to undergo screening as compared to 55% of controls. All comparisons were statistically significant. Conclusion Compared to controls, PC family members perceive their personal PC risk to be high and report a higher rate of PC-related cancer worry/concern. A high percentage of family members and controls indicate receptivity to PC screening, especially if it were a blood test. If the screening is more invasive (i.e. EUS), interest declines, particularly among controls. This information can help investigators to understand and address potential motivators and barriers to recruitment for future PC studies and indicates that there is a need to develop educational interventions among these families to reduce their actual and perceived pancreatic cancer risk. Citation Information: Cancer Prev Res 2008;1(7 Suppl):B24.

Abstract B128 Background The B vitamin folate affects colorectal carcinogenesis through effects on nucleotide synthesis and possibly DNA methylation. Animal models and epidemiologic evidence suggest that folate from diet and supplements can prevent the development of colorectal cancer (CRC), but that high folate intake after adenomas or cancer are established may accelerate progression or recurrence. The chemotherapeutic agent 5-fluorouracil inhibits thymidylate synthase, a folate-metabolizing enzyme, and the effect of folate intake on treatment efficacy is unknown. Supplement use and folate intake are high among adults in the U.S., particularly with food supply fortification, and higher still among cancer patients and survivors. However, few studies have addressed use of folic acid-containing supplements (FAS) among colorectal cancer patients, how use changes after diagnosis, and what factors determine changes in FAS use. Methods The Colon Cancer Family Registry (CCFR) is a multicenter study of colorectal cancer cases, their family members and controls, recruited since 1998 at six sites in the United States, Australia and Canada. The current analysis includes 1,092 CRC cases with epidemiologic data available from questionnaires administered at enrollment, asking about supplement use and other risk factors before diagnosis, and from follow-up questionnaires about 5 years later. Baseline characteristics for cases who began using FAS (including multivitamins) after diagnosis were compared to those for cases who used FAS neither before nor after diagnosis. We used logistic regression models to evaluate associations between age, sex, CCFR site, race, education, income, lifetime exercise, smoking, body mass index, and diet and change in FAS use, adjusting each model for age, sex and site, when appropriate. Results FAS use before CRC diagnosis was 35.4%, while 55.1% of cases used FAS after diagnosis. Women were more likely to begin FAS use after diagnosis (OR 1.60, 95% CI 1.17-2.19). Current smokers were less likely than nonsmokers to begin FAS use (OR 0.63, 95% CI 0.40-0.97), as were those consuming more red meat (OR 0.38, 95% CI 0.21-0.70 for those in the highest versus lowest intake categories, ptrend=0.012). Subjects with higher fruit intake were more likely to begin FAS use (OR 1.95, 95% CI 1.16-3.30 for highest versus lowest intake, ptrend=0.013). We also observed a suggestive association between lifetime physical activity and change in FAS use for active versus less active subjects (OR 1.48, 95% CI 1.01-2.16), though there was no trend with increasing levels of physical activity (ptrend=0.27). Finally, we found that residents of non-U.S. countries were less likely to begin FAS use (OR 0.57, 95% CI 0.33-0.96 for Ontario, Canada and OR 0.21, 95% CI 0.11-0.38 for Australia). Conclusions: Our analysis showed substantial increases in the use of folic acid-containing supplements after diagnosis with colorectal cancer in CCFR participants, especially among women, U.S. residents, nonsmokers, and those who consumed more fruit and less meat. This study begins to characterize CRC patients likely to be using FAS, and suggests that its use is widespread. This finding is notable given evidence that folate may accelerate progression of colorectal cancer, and the unknown effect that FAS intake may have on the efficacy of cancer treatment and on survival. Citation Information: Cancer Prev Res 2008;1(7 Suppl):B128.

Abstract A61 Background Patients with Barrett's esophagus (BE) and dysplasia are ideal candidates for chemopreventive strategies to reduce cancer risk. Difluoromethylornithine (DMFO) is a promising chemopreventive agent that inhibits polyamine synthesis involved in the regulation of cell growth and differentiation. We conducted a pilot study to examine the effects of DFMO on esophageal histopathology, mucosal polyamine content, and gene expression. Materials and Methods Patients (n=10) with histologically confirmed BE and low grade dysplasia were enrolled in a single-arm study of DFMO (0.5 g/m2/d) given continuously for 6 months. Esophagoscopy with biopsies (4 quadrants every 1 cm) was performed at baseline, 3 and 6 months, and at 12 months (where available). Grading of dysplasia was performed by a GI pathologist blinded to clinical/biomarker data. Audiology was also performed. Mucosal polyamine content was measured by high-performance liquid chromatography, and analyzed using paired t-tests. Gene expression profiling was performed using a cDNA two-color chip, and was compared between baseline and 6 months. Expression data was normalized using the two-channel fastlo routine and differential expression was computed using a Wilcoxon signed-rank test. Results After 6 months of DFMO treatment, histopathological response data demonstrate regression of dysplasia (n= 1), stable disease (n= 8), and progression of dysplasia (n= 1). Within the stable category, 2 patients with extensive low grade dysplasia showed only focal dysplasia based upon 4 or more biopsies at 6 months. Histolopathological improvements seen at 6 months were maintained at 12 months. Overall, DFMO was well tolerated. One patient had subclinical, unilateral ototoxicity. Modulation of polyamines (spermine, spermidine, putrescine, histamine) was shown only for putrescine which decreased with treatment (6 mo., p=0.07) and showed recovery to baseline at 12 months (p= 0.01). Gene expression profiling revealed modulated genes of potential biological importance (KLF5, KAL1, RFC5, TIMM8a, TC10) and of these, Krüppel-like factor 5 (KLF5) was downregulated whereas the others were upregulated. Conclusion >Further study of the chemopreventive effect of DFMO upon BE with dysplasia is suggested. DFMO treatment suppressed mucosal putrescine levels and downregulated KLF5, an important mediator of cell proliferation. Citation Information: Cancer Prev Res 2008;1(7 Suppl):A61.