K-medoids clustering of hospital admission characteristics to classify severity of influenza virus infection.

Background: Patients are admitted to the hospital for respiratory illness at different stages of their disease course. It is important to appropriately analyse this heterogeneity in surveillance data to accurately measure disease severity among those hospitalized. The purpose of this study was to determine if unique baseline clusters of influenza patients exist and to examine the association between cluster membership and in-hospital outcomes. Methods: Patients hospitalized with influenza at two hospitals in Southeast Michigan during the 2017/2018 (n = 242) and 2018/2019 (n = 115) influenza seasons were included. Physiologic and laboratory variables were collected for the first 24 h of the hospital stay. K-medoids clustering was used to determine groups of individuals based on these values. Multivariable linear regression or Firth's logistic regression were used to examine the association between cluster membership and clinical outcomes. Results: Three clusters were selected for 2017/2018, mainly differentiated by blood glucose level. After adjustment, those in C171 had 5.6 times the odds of mechanical ventilator use than those in C172 (95% CI: 1.49, 21.1) and a significantly longer mean hospital length of stay than those in both C172 (mean 1.5 days longer, 95% CI: 0.2, 2.7) and C173 (mean 1.4 days longer, 95% CI: 0.3, 2.5). Similar results were seen between the two clusters selected for 2018/2019. Conclusion: In this study of hospitalized influenza patients, we show that distinct clusters with higher disease acuity can be identified and could be targeted for evaluations of vaccine and influenza antiviral effectiveness against disease attenuation. The association of higher disease acuity with glucose level merits evaluation.

Distinct influenza surveillance networks and their agreement in recording regional influenza circulation: Experience from Southeast Michigan.

INTRODUCTION: In Southeast Michigan, active surveillance studies monitor influenza activity in hospitals, ambulatory clinics, and community households. Across five respiratory seasons, we assessed the contribution of data from each of the three networks towards improving our overall understanding of regional influenza circulation. METHODS: All three networks used case definitions for acute respiratory illness (ARI) and molecularly tested for influenza from research-collected respiratory specimens. Age- and network-stratified epidemic curves were created for influenza A and B. We compared stratified epidemic curves visually and by centering at seasonal midpoints. RESULTS: Across all seasons (from 2014/2015 through 2018/2019), epidemic curves from each of the three networks were comparable in terms of both timing and magnitude. Small discrepancies in epidemics recorded by each network support previous conclusions about broader characteristics of particular influenza seasons. CONCLUSION: Influenza surveillance systems based in hospital, ambulatory clinic, and community household settings appear to provide largely similar information regarding regional epidemic activity. Together, multiple levels of influenza surveillance provide a detailed view of regional influenza epidemics, but a single surveillance system-regardless of population subgroup monitored-appears to be sufficient in providing vital information regarding community influenza epidemics.

Multi-institutional phase I study of low-dose ultra-fractionated radiotherapy as a chemosensitizer for gemcitabine and erlotinib in patients with locally advanced or limited metastatic pancreatic cancer.

PURPOSE: Gemcitabine (G) has been shown to sensitize pancreatic cancer to radiotherapy but requires lower doses of G and thus delays aggressive systemic treatment, potentially leading to distant failure. We initiated a phase I trial combining ultra-fractionated low-dose radiotherapy with full dose G and erlotinib in the treatment of patients with advanced pancreatic cancer. METHODS: Patients with locally advanced or metastatic pancreatic cancer confined to the abdomen and an ECOG performance status (PS) of 0-1 who had received 0-1 prior regimens (without G or E) and no prior radiotherapy were eligible. Patients were treated in 21 day cycles with G IV days 1 & 8, E once PO QD, and twice daily RT fractions separated by at least 4h on days 1, 2, 8, and 9. Whole abdominal RT fields were used. Primary endpoint was to define dose limiting toxicity (DLT) and the maximum tolerated dose (MTD). RESULTS: 27 patients (median age 64 years and 15 male) were enrolled between 11/24/08 and 4/12/12. 1 patient withdrew consent prior to receiving any protocol therapy. 17 patients had a PS of 1. The majority of patients were stage IV. One DLT was noted out of 7 patients at dose level (DL) 1. Subsequently no DLTs were noted in 3 patients each enrolled at DL2-4 or 11 patients in the expansion cohort. The majority of grade 3 toxicities were hematologic with 1 grade 5 bowel perforation in dose level 1 in cycle 4. Best response in 24 evaluable patients: PR (8), stable (15), PD 1. Median survival for the entire group was 9.1 months. CONCLUSION: This phase I study combining low-dose ultra-fractionated RT as a sensitizer to full dose G plus E was well tolerated with encouraging efficacy. This represents a novel strategy worthy of further investigation in advanced pancreatic cancer patients.

Limitations of the bowel bag contouring technique in the definitive treatment of cervical cancer.

PURPOSE: Incidence of acute grade 3 and 4 small bowel toxicity in the definitive treatment of cervical cancer is approximately 15%. Given uncertainties in position of the bowel at time of treatment, techniques including the contouring of a bowel bag have been suggested. The purpose of this study is to describe interfraction variability in bowel location for the female pelvis with intact reproductive organs and to characterize the ability of the bowel bag technique, as described in the Radiation Therapy Oncology Group pelvic normal tissue contouring guidelines, to account for organ motion in this specific clinical setting. METHODS AND MATERIALS: Bowel position was assessed for 45 computed tomographic scans used in treatment planning for 9 consecutive cervical cancer patients. After a single operator contoured bowel loops, most superior, anterior, posterior, and inferior positions of bowel were recorded. Mixed effects models were used to assess significance of interfraction variability. Frequency of bowel loop migration outside of the bowel bag was then considered for each patient given all potential bowel bag volumes. Standardized scoring was used to determine additional margins that would be required to account for 95%, 90%, and 85% of significant bowel motion. RESULTS: Interfraction variability in the inferior-most bowel position was significant (P = .002). Median maximum variation in the inferior bowel position was 2.1 cm (range, 0.9 cm-4.8 cm). When applying the bowel bag technique, 100% of bowel motion was accounted for as the bowel translated laterally, anteriorly, posteriorly, and superiorly, though accounted for just 70.3% of motion in the inferior direction. A 4-cm inferior margin was required to account for 90% of motion in the inferior direction. CONCLUSIONS: In the intact female pelvis, the bowel bag technique is successful in accounting for most interfraction variability in bowel position but underestimates inferior motion. Until an improved approach to predicting small bowel motion can be routinely implemented, a focus on decreasing dose to potential bowel space should be emphasized.

Correlation of admissions statistics to graduate student success in medical physics.

The purpose of this work is to develop metrics for evaluation of medical physics graduate student performance, assess relationships between success and other quantifiable factors, and determine whether graduate student performance can be accurately predicted by admissions statistics. A cohort of 108 medical physics graduate students from a single institution were rated for performance after matriculation based on final scores in specific courses, first year graduate Grade Point Average (GPA), performance on the program exit exam, performance in oral review sessions, and faculty rating. Admissions statistics including matriculating program (MS vs. PhD); undergraduate degree type, GPA, and country; graduate degree; general and subject GRE scores; traditional vs. nontraditional status; and ranking by admissions committee were evaluated for potential correlation with the performance metrics. GRE verbal and quantitative scores were correlated with higher scores in the most difficult courses in the program and with the program exit exam; however, the GRE section most correlated with overall faculty rating was the analytical writing section. Students with undergraduate degrees in engineering had a higher faculty rating than those from other disciplines and faculty rating was strongly correlated with undergraduate country. Undergraduate GPA was not statistically correlated with any success metrics investigated in this study. However, the high degree of selection on GPA and quantitative GRE scores during the admissions process results in relatively narrow ranges for these quantities. As such, these results do not necessarily imply that one should not strongly consider traditional metrics, such as undergraduate GPA and quantitative GRE score, during the admissions process. They suggest that once applicants have been initially filtered by these metrics, additional selection should be performed via the other metrics shown here to be correlated with success. The parameters used to make admissions decisions for our program are accurate in predicting student success, as illustrated by the very strong statistical correlation between admissions rank and course average, first year graduate GPA, and faculty rating (p < 0.002). Overall, this study indicates that an undergraduate degree in physics should not be considered a fundamental requirement for entry into our program and that within the relatively narrow range of undergraduate GPA and quantitative GRE scores of those admitted into our program, additional variations in these metrics are not important predictors of success. While the high degree of selection on particular statistics involved in the admissions process, along with the relatively small sample size, makes it difficult to draw concrete conclusions about the meaning of correlations here, these results suggest that success in medical physics is based on more than quantitative capabilities. Specifically, they indicate that analytical and communication skills play a major role in student success in our program, as well as predicted future success by program faculty members. Finally, this study confirms that our current admissions process is effective in identifying candidates who will be successful in our program and are expected to be successful after graduation, and provides additional insight useful in improving our admissions selection process.

Women at increased risk for cardiac toxicity following chemoradiation therapy for esophageal carcinoma.

PURPOSE: The purpose of this study was to identify factors associated with cardiac toxicity in patients treated with chemoradiation therapy (CRT) for esophageal carcinoma. METHODS AND MATERIALS: One hundred twenty-seven patients with adenocarcinoma or squamous cell carcinoma of the esophagus treated from July 2002 to June 2011 at 2 academic institutions with preoperative or definitive CRT were retrospectively reviewed. Association of cardiac toxicity with a number of variables was investigated, including heart disease, cardiac bypass and angioplasty, diabetes, insulin use, smoking, chemotherapy regimen, and tumor location. T test assessed risk of cardiac toxicity secondary to age. Dose volume histograms (DVH) were evaluated for percentage of heart volume receiving >20, 30, 40, and 50 Gy (V20-V50). The Fisher exact test analyzed for an association between dose volume parameters and cardiac toxicity. RESULTS: Patient population included 100 men and 27 women with a mean age of 64 years. Median follow-up was 12.7 months (range, 0.3-99.6 months). Any cardiac toxicity occurred in 28 patients, the majority of which were pericardial effusion (23/28). Odds ratio for toxicity in women was 4.15 (95% confidence interval [CI], 1.63-10.50; P = .0017) and time to cardiac toxicity by sex was significant (P = .0003). Patients above the median cutoff for V20, V30, and V40 had increased odds of developing cardiac toxicity (P = .03, .008, .002). There was 4.0 increased odds of developing cardiac toxicity with V40 >57% (95% CI, 1.5-10.3, P = .002). On multivariable logistic regression analysis, sex was the only variable associated with any cardiac toxicity and pericardial effusion (P = .0016, P = .0038). None of the other investigated variables were associated with increased risk of cardiac toxicity. CONCLUSIONS: Female patients and dose greater than the median for V20-V40 were associated with the development of cardiac toxicity, specifically pericardial effusion. These data suggest exercising increased care when designing radiation fields in women undergoing CRT for esophageal carcinoma, as pericardial effusion may be a long-term complication.

Inflammatory breast cancer in north Africa: comparison of clinical and molecular epidemiologic characteristics of patients from Egypt, Tunisia, and Morocco.

Understanding molecular characteristics that distinguish inflammatory breast cancer (IBC) from non-IBC is crucial for elucidating breast cancer etiology and management. We included 3 sets of patients from Egypt (48 IBC and 64 non-IBC), Tunisia (24 IBC and 40 non-IBC), and Morocco (42 IBC and 41 non-IBC). Egyptian IBC patients had the highest combined erythema, edema, peau d'orange, and metastasis among the 3 IBC groups. Egyptian IBC tumors had the highest RhoC expression than Tunisians and Moroccan IBCs (87% vs. 50%, vs. 38.1, for the 3 countries, respectively). Tumor emboli were more frequent in Egyptian IBC than non-IBC (Mean ± SD: 14.1 ± 14.0 vs. 7.0 ± 12.9, respectively) (P < 0.001) and Tunisians (Mean ± SD: 3.4 ± 2.5 vs. 1.9 ± 2.0, respectively) (P < 0.01). There was no difference of emboli in Moroccan tumors (1.7 ± 1.2 vs. 1.8 ± 1.2 for IBC and non-IBC, respectively (P=0.66). This study illustrates that RhoC overexpression and tumor emboli are more frequent in IBC relative to non-IBC from Egypt and Tunisia. Tumors of Moroccans were significantly different from Egyptian and Tunisian tumors for RhoC expression and emboli. Future studies should focus on relating epidemiologic factors and clinical pictures to molecular features of IBC in these and other populations.