Faculty Bibliography

Background: Case reports suggest that NS is responsive to dietary modifications including a gluten-free diet (GFD). In celiac disease, zonulin is released from enterocytes after exposure to gliadin, activates protease activated receptor 2 (PAR2), and perturbs the actin cytoskeleton and cell-cell junctions in the gut. PAR2 is present on podocytes and, therefore, zonulin may increase glomerular permeability in NS. We conducted this study to test the hypothesis that plasma zonulin levels are elevated in pediatric patients with NS.
Method(s): Plasma specimens collected from patients <=18 yr old with minimal change disease or FSGS enrolled in the NEPTUNE study, were tested. Clinical and laboratory data were retrieved coincident with the visit when the zonulin level was measured. Samples were available for testing from the 4 or 8 month visit. Plasma zonulin levels were measured by ELISA. Results (mean+/-SD or median (IQR)) were analyzed by t-test, Wilcoxon, Kruskal-Wallis, or linear regression and considered significant if P<0.05 Results: There were 113 patients, 9.5+/-4.9 yr, 53% male, 42% white, 40% black and 18% other. Disease classification was infrequent relapser in 27%, frequent relapser/steroid dependent 42% and steroid resistant 30%. The mean BP, eGFR, and serum albumin were normal. Urine protein:creatinine (UPC) ratio was 3.9+/-6.9 (g:g). The plasma zonulin level in NS children was 14.2+/-6.0 vs 10+/-2.5 ng/ml in healthy adults (P<0.01) and was >3 standard deviations above the mean in 27%. There was a trend toward lower zonulin levels in children with UPC >=2 vs <2, 12.9(7.4) vs 16.7(8.0) (P=0.051). Plasma zonulin levels did not differ by eGFR, disease classification, or BP. Plasma zonulin and serum albumin concentrations were directly correlated, r=0.24, P=0.04.
Conclusion(s): The plasma zonulin level was significantly elevated in more than a quarter of children with NS and was unrelated to BP or eGFR. We observed a significant relationship between zonulin values and serum albumin but not proteinuria. There was a trend to lower zonulin levels in children with nephrotic-range proteinuria. Further study is needed to determine the relationship between plasma zonulin levels and proteinuria and to test whether the plasma zonulin level can be used to predict response to a GFD in children with NS

Background: Although the optimal timing of adjuvant chemotherapy (AC) for stage III colon cancer patients has been debated, most studies recommend initiating AC within approximately 60 days of surgery. Significant disparities in timeliness of AC initiation in colon cancer have been reported in public versus private hospitals, with longer time to AC at public hospitals. We evaluated whether timeliness of AC differed between a public and a private hospital, both affiliated with the same major academic institution in New York City. Methods: We conducted a retrospective cohort study of Stage III colon cancer patients who underwent surgery and received AC at the same institution from 2008-2015 at NYU Langone Medical Center's affiliated public hospital (Bellevue) or its private hospital (Tisch). Patient data were obtained through review of hospital tumor registry and electronic medical records. Patient characteristics were compared by hospital. We defined timeliness as receipt of AC within 60 days postoperatively. Univariate and stepwise multivariable logistic regressions were used to identify factors associated with timely AC. Results: Forty three patients at Bellevue Hospital and 79 patients at Tisch Hospital who underwent surgery and received AC at the same institution were included. Median number of days to AC was significantly greater among patients receiving care at Bellevue (53, range 31-231) compared to Tisch (43, range 25-105; p=0.002). However, the percentage of patients who received timely AC did not differ substantially at Bellevue and Tisch (74% vs 81%, p=0.40). Individual characteristics significantly associated with timely initiation of AC were non-Hispanic ethnicity (OR: 2.71, 95% CI: 1.06-6.95), married (OR: 2.89, 95%CI: 1.15-7.30), and laparoscopic (vs open) surgery (OR: 4.30, 95%CI: 1.64-11.25). The odds of receiving timely AC at Bellevue compared to Tisch was not significant (OR: 0.68, 95% CI: 0.28-1.65). When hospital and other factors were examined jointly, only age (OR: 0.95/year, 95% CI: 0.91-0.99) and laparoscopic (vs open) surgery (OR: 5.65, 95% CI: 1.92-16.62) remained as important factors associated with receiving timely AC (Likelihood Ratio Chi-Square=14.95, p=0.0019). When hospital was omitted from multivariable analysis, age and surgery type still remained the only significant factors associated with timely AC (OR's unchanged, Likelihood Ratio Chi-Square=14.81, p-value=0.0006). Conclusions: The proportion of patients receiving timely AC within 60 days of surgery was similar at both an affiliated public and private hospital at NYU Langone Medical Center. Age and type of surgery were significant predictors of timeliness in our population. Further research should be conducted to understand how system-level factors may promote timely receipt of care

Patients with multifocal breast cancers (MBCs) have a poorer prognosis than patients with unifocal breast cancers. Studies have attributed this to tumor size underestimation in MBC. An alternative hypothesis is that some MBCs behave in a fashion analogous to the "satellite" and "in-transit metastasis" observed in melanoma and, thereby, are more clinically aggressive. We identified 79 cases of MBC, which we classified into 2 groups: study cases defined as >/=2 morphologically similar tumor foci with >/=1 focus without in situ carcinoma (n = 21); and a control group defined as >/=2 morphologically similar or dissimilar foci with associated in situ carcinoma in all foci (n = 58). The odds of being a study case is 1.86 (95% confidence interval [CI] 1.26-2.74) times greater per unit increase in number of tumor foci (median of 4 tumor foci; P = .002). Study cases were 73.33 (95% CI = 8.91-603.16) times more likely to have lymphovascular invasion (LVI) and 14.72 (95% CI = 4.37-49.61) times more likely to have nodal metastases. Grade I/II tumors were 0.20 (95% CI = 0.07-0.59) times less likely to be study cases. There was a significant positive interaction (P < 0.001) indicated by the relationship of LVI status and nodal status with the study case and control group. We conclude that there is a subset of MBC that presents with more numerous tumor foci and a higher rate of nodal metastasis. The aggressive behavior of these cases may be attributed to their proclivity for LVI.

OBJECTIVES: To evaluate the efficacy and safety of irinotecan and bevacizumab in recurrent ovarian cancer. The primary objective was to estimate the progression free survival (PFS) rate at 6months. Secondary objectives included estimation of overall survival (OS), objective response rate (ORR), duration of response, and an evaluation of toxicity. METHODS: Recurrent ovarian cancer patients with no limit on prior treatments were eligible. Irinotecan 250mg/m2 (amended to 175mg/m2 after toxicity assessment in first 6 patients) and bevacizumab 15mg/kg were administered every 3weeks until progression or toxicity. Response was assessed by RECIST or CA-125 criteria every 2cycles. RESULTS: Twenty nine patients enrolled (10 were platinum-sensitive and 19 were platinum-resistant). The median number of prior regimens was 5 (range 1-12); 13 patients had prior bevacizumab and 11 prior topotecan. The PFS rate at 6months was 55.2% (95% CI: 40%-77%). The median number of study cycles given was 7 (range 1-34). Median PFS was 6.8months (95% CI: 5.1-12.1months); median OS was 15.4months (95% CI: 11.9-20.4months). In this study, no complete response (CR) was observed. The objective response rate (ORR; PR or CR) for all patients entered was 27.6% (95% CI: 12.7%-47.2%) and the clinical benefit rate (CR+PR+SD) was 72.4% (95% CI: 52.8%-87.3%); twelve patients experienced duration of response longer than 6months. In the 24 patients with measurable disease, a partial response (PR) was documented in 8 (30%) patients; 13 patients maintained stable disease (SD) at first assessment. The most common grade 3/4 toxicity was diarrhea. No treatment-related deaths were observed. CONCLUSIONS: Irinotecan and bevacizumab has activity in heavily pre-treated patients with recurrent ovarian cancer, including those with prior bevacizumab and topoisomerase inhibitor use.

Myelofibrosis is a chronic and progressive myeloproliferative neoplasm characterized by anemia, splenomegaly, debilitating symptoms and leukemic transformation. Ruxolitinib, an oral JAK1/2 inhibitor, is highly effective in ameliorating systemic symptoms and reducing splenomegaly. Current clinical research is focused on the evaluation of agents based on pre-clinical rationale that can result in disease course modification. Panobinostat is a pan-histone deacetylase inhibitor that has demonstrated clinical activity as a single agent in early phase trials of myelofibrosis. We previously conducted a phase I trial of panobinostat monotherapy in patients with myelofibrosis and determined 25mg thrice weekly as the recommended phase II dose. We then completed an investigator initiated, Simon 2-stage, phase II trial of 22 myelofibrosis patients at our single institution. After 6 cycles of therapy, the overall response rate by IWG-MRT criteria was 36% (8/22; 95% CI: 16-56%). The median percent reduction in spleen volume was 34% (range, 1.6%-73%) in eight evaluable patients. The average reduction in JAK2V617F allele burden was 6.8% (Range; -4.0% to 20.2%) and one patient obtained a complete molecular response. Six patients remained on therapy in the extension phase for a median of 18 months (range, 7-44). Treatment discontinuation was frequent due to patient/physician perception of therapy ineffectiveness. The optimal dosing of panobinostat for the treatment of MF remains somewhat ill-defined but appears to be most effective and better tolerated when administered at lower doses over a prolonged duration of therapy.

Purpose: To determine feasibility and explore the clinical efficacy of concurrent radiotherapy and carboplatin as adjuvant treatment of triple negative breast cancer (TNBC). Patients and Methods: Women with Stage I-II TNBC were treated after surgery in a phase I-II prospective trial [NCT01289353]. Weekly carboplatin (AUC = 2.0) was delivered for 6 weeks. Concurrent radiotherapy was delivered in the prone position during weeks 2-4, for a total dose of 40.5 Gy in 15 fractions to the breast, and 46.5 Gy in 17 fractions to the tumor bed. Adverse events (AE) were assessed weekly during treatment, once at 45-60 d, and every 6 mo thereafter, using the Common Terminology Criteria for AE (CTCAE) v3.0. Results: A total of 39 patients accrued and 36 received treatment. Eight patients (22%, exact 95% CI: 10%, 39%) developed grade 2 or greater acute radiation dermatitis. Overall, grade 2 AE were seen in nine and grade 3 in two patients. Twenty-three patients (64%) received additional adjuvant chemotherapy. With a median follow-up of 48 mo, 34/36 (94%) are alive and disease free. One patient died of pulmonary failure with possible but unproven breast cancer recurrence, and one patient died of pelvic malignancy. One patient recurred locally and is alive and disease free after surgical management. Brisk lymphocytic infiltrate was present pre-treatment in 39% of 18 patients with evaluable tumor. Conclusions: Adjuvant concurrent carboplatin and prone accelerated radiotherapy is a well-tolerated and promising treatment of early stage TNBC. The observed 3% compares favorably with the expected 30% recurrence rate within 1-4 y from treatment, warranting further studies.

PURPOSE: Maximum dose to the left anterior descending artery (LADmax) is an important physical constraint to reduce the risk of cardiovascular toxicity. We generated a simple algorithm to guide the positioning of the tangent fields to reliably maintain LADmax <10 Gy. METHODS AND MATERIALS: Dosimetric plans from 146 consecutive women treated prone to the left breast enrolled in prospective protocols of accelerated whole breast radiation therapy, with a concomitant daily boost to the tumor bed (40.5 Gy/15 fraction to the whole breast and 48 Gy to the tumor bed), provided the training set for algorithm development. Scatter plots and correlation coefficients were used to describe the bivariate relationships between LADmax and several parameters: distance from the tumor cavity to the tangent field edge, cavity size, breast separation, field size, and distance from the tangent field. A logistic sigmoid curve was used to model the relationship of LADmax and the distance from the tangent field. Furthermore, we tested this prediction model on a validation data set of 53 consecutive similar patients. RESULTS: A lack of linear relationships between LADmax and distance from cavity to LAD (-0.47), cavity size (-0.18), breast separation (-0.02), or field size (-0.28) was observed. In contrast, distance from the tangent field was highly negatively correlated to LADmax (-0.84) and was used in the models to predict LADmax. From a logistic sigmoid model we selected a cut-point of 2.46 mm (95% confidence interval, 2.19-2.74 mm) greater than which LADmax is <10 Gy (95% confidence interval, 9.30-10.72 Gy) and LADmean is <3.3 Gy. CONCLUSIONS: Placing the edge of the tangents at least 2.5 mm from the closest point of the contoured LAD is likely to assure LADmax is <10 Gy and LADmean is <3.3 Gy in patients treated with prone accelerated breast radiation therapy.

We consider the non-inferiority (or equivalence) test of the odds ratio (OR) in a crossover study with binary outcomes to evaluate the treatment effects of two drugs. To solve this problem, Lui and Chang (2011) proposed both an asymptotic method and a conditional method based on a random effects logit model. Kenward and Jones (1987) proposed a likelihood ratio test (LRTM ) based on a log linear model. These existing methods are all subject to model misspecification. In this paper, we propose a likelihood ratio test (LRT) and a score test that are independent of model specification. Monte Carlo simulation studies show that, in scenarios considered in this paper, both the LRT and the score test have higher power than the asymptotic and conditional methods for the non-inferiority test; the LRT, score, and asymptotic methods have similar power, and they all have higher power than the conditional method for the equivalence test. When data can be well described by a log linear model, the LRTM has the highest power among all the five methods (LRTM , LRT, score, asymptotic, and conditional) for both non-inferiority and equivalence tests. However, in scenarios for which a log linear model does not describe the data well, the LRTM has the lowest power for the non-inferiority test and has inflated type I error rates for the equivalence test. We provide an example from a clinical trial that illustrates our methods