Faculty Bibliography

Introduction: Edema is a frequent clinical observation after chemoradiation treatment (CRT) for oral/oropharyngeal cancer (O/OP Ca). Our aims were to reliably quantify edema from video fluoroscopy (VF) at 3 time points (baseline 1-month (mo) and 4-mo post CRT) and to explore the relationship between edema and (a) patient-reported outcomes (EAT-10) and (b) functional impairment on VF (Dynamic Imaging Grade of Swallowing Toxicity DIGEST).
Material(s) and Method(s): 15 patients (7 M; age 38-76) with O/OP Ca received radiotherapy (70 Gy 7 weeks) and 3 weekly doses of cisplatin. VF was completed pre-CRT 1-mo and 4-mo post-CRT. Edema was captured by measuring posterior pharyngeal wall (PPW) thickness and pharyngeal area (PA) at rest. EAT-10 surveys were completed on the day of VF. DIGEST scores were rated according to published protocols. Mixed model repeated measures ANOVAs were run for each edema measure (PPW PA) to test for the effect of TIME EAT-10 and DIGEST while controlling for age and sex.
Result(s): For PPW we found a main effect of TIME but not EAT-10 or DIGEST (Table 1). Post-hoc comparisons revealed a significant worsening from mean at baseline (4.1 mm) to 4-mo post CRT (6.0 mm) but not at 1-mo post CRT (5.4 mm). For PA we found a main effect of TIME and of DIGEST grade (Table 2). Mean PA was significantly smaller at 1-mo post CRT (527 mm2) compared with baseline (716 mm2) but not different from 4-mo post CRT (652 mm2). Mean PA was significantly greater for grade 2 (751 mm2) compared with grade 0 (442 mm2) contrary to the hypothesized direction.
Conclusion(s): The data confirm that post-CRT edema can be quantified on 2D lateral VF. Patient reported outcomes (EAT-10) were not independently predictive of edema. Surprisingly worse DIGEST grades were associated with increased pharyngeal area at rest perhaps reflecting impairment associated with pharyngeal atrophy not edema. Future work should monitor patients' edema and swallow function over a longer time period and at a greater frequency

Introduction: Videofluoroscopic swallow studies (VFSS) are commonly requested by otolaryngologists to evaluate swallow function in patients with cough but no data exists on the utility of VFSS in this population. We aim to determine which history clinical exam and laryngoscopy findings correlate with abnormal VFSS findings in patients with cough.
Material(s) and Method(s): Ten items from flexible videolaryngoscopy were recorded including: motion abnormalities pooling of secretions pharyngeal asymmetry and glottal insufficiency. VFSS findings recorded included penetration aspiration delayed initiation and presence of residue after swallow. Nonparametric statistical analysis was performed to determine correlations between history and clinical exam observation and VFSS findings. A total of 405 patients with a chief complaint of cough were referred to speech language pathology. Of those 107 had a VFSS and 93 had an esophagram. Forty-five patients had a VFSS after referral and were included in the analysis.
Result(s): Age (p = 0.35) glottal insufficiency (p = 0.33) pooling of secretions (p = 0.10) any videolaryngoscopy abnormality (p = 0.07) cardiopulmonary history (p = 0.29) and other variables did not correlate VFSS abnormalities in patients with cough. Only gender (p = 0.02) was a predictor of an abnormal VFSS (86% males and 57% females).
Conclusion(s): This study found that videolaryngoscopy and clinical exam were not predictive of VFSS abnormalities in patients with cough

Purpose: Despite being widely adopted in clinical practice, the rehabilitative potential of swallowing exercises is not well documented (Langmore & Pisegna, 2015). While collecting a sample of 98 videofluoroscopies (VF) from healthy individuals to serve as a control group for an unrelated study, we discovered 13 subjects with incidental findings of impaired swallowing (safety and/or efficiency deficits). Our purpose was to explore the impact of a standardized 'one-size-fits-all' treatment (tx) protocol on impaired swallowing function in this cohort of otherwise healthy individuals. Method(s): 13 healthy individuals (9 F, mean age = 71.5, SD = 11.9) completed 8 weeks of swallowing exercises. Treatment sessions (once per week) consisted of 20 repetitions of each of the following exercises: effortful swallows, tongue hold swallows, supraglottic swallows, Shaker exercises and Mendelsohn maneuvers, as well as 10 repetitions of effortful pitch glides. Subjects were also asked to complete daily homework consisting of 3 additional treatment sets per day. VF was collected pre-and post-tx with a standardized protocol and scored using the MBSimPTM method. Scores for components 1-5 and 6-16 were combined for an oral total (OT) and pharyngeal total (PT) respectively. Wilcoxon rank sum tests compared OT and PT scores from pre-to post-tx. Result(s): The pre-tx and post-tx OT median scores remained unchanged (4). The median PT score was 10 pre-tx (range 2-14) and reduced to 7 post-tx (range 3-11), though this change narrowly missed statistical significance (Z =-1.99; p = .058). Post-hoc evaluations revealed that 8 subjects demonstrated improved PT scores, 2 worsened, 3 were unchanged and that the greatest changes came from components 6 (initiation of the pharyngeal swallow), 8 (laryngeal elevation), 15 (tongue base retraction) and 16 (pharyngeal residue). Conclusions (Including Clinical Relevance): Our sample of otherwise healthy individuals with VF evidence of impaired swallowing completed a standardized 'one-size-fits-all' approach to dysphagia rehabilitation that is common place in clinical practice. The approach appeared to rehabilitate aspects of swallowing function-especially in the pharyngeal phase-for the majority of subjects. Future research should compare physiologically-targeted exercises with one-size-fits all approaches. Further investigations into dose, frequency and maintenance of exercise interventions will be vital contributions

Purpose: Non-tuberculous mycobacteria (NTM) is an intracellular microorganism that causes cavitary disease and nodular bronchiectatic disease of the lung. Common symptoms include chronic cough, sputum production and frequent pneumonias. Patients with NTM appear to have impaired swallow function as represented by impaired airway protection. Our aim is to determine bolus clearance in patients with NTM compared with a control. Method(s): Videofluoroscopy (VF) was prospectively collected from 195 patients:132 patients with NTM (90 females; ages 30-90) and 63 age-and sex-matched normal controls with normal pulmonary function tests and no pulmonary disease (39 females; ages 27-92).Two boluses of 3, 5, 10 mL thin liquid, two 5 mL puree Varibar, and cracker were analyzed per subject (N = 1,755 swallows).Outcome measures included ordinal ratings of residue in the valleculae and pyriform sinuses. The correlation between clinical information and the present of pharyngeal residue structures was analyzed by Fisher's exact test for categorical variables and Wilcoxon's rank sum test for comparisons of continuous variables between groups. Result(s): Inter-and intra-rater reliability of the ordinal ratings were assessed using two-way random intraclass correlation coefficients (ICC(2,1)) on 20% of the data with good results (intra-rater: ICC = 0.92, 95% CI 0.70-0.97 and inter-rater: ICC = 0.92, 95% CI 0.81-0.97).The ratings of residue in the valleculae were significantly higher in the NTM group compared to the control group for one of the 3 mL bolus (p = 0.008), for 5 mL boluses (p = 0.009 and p = 0.004), 10 mL boluses (p = 0.0005 and p = 0.0006), puree (p = 0.006), cracker (p = 0.005).The ratings of residue in the pyriform sinuses were also significantly higher in the NTM group for 3 mL boluses (p = 0.0002 and p = 0.01), 5 mL boluses (p = 0.0003 and p = 0.002), 10 mL boluses (p = 0.001 and p = 0.001), puree trials (p<0.0001 and p<0.0001).Wilcoxon's rank sum test determined no age difference and Fisher's exact test determined no gender difference between the NTM and control groups. Conclusions (Including Clinical Relevance): Patients with NTM appear to have reduced bolus clearance than healthy individuals with no pulmonary disease, as represented by ratings of residue in the valleculae and pyriform sinuses.Future work is needed to elucidate the interaction between the respiratory-swallowing systems and airway protection and responsiveness to swallowing treatment for patients with NTM

BACKGROUND: There is no effective therapy for patients (pts) with IDH-mutant gliomas that progress after RT and chemotherapy. At time of progression, these tumors have often transformed to glioblastoma (GBM) and have increased numbers of somatic mutations, i.e. have a ?hypermutator phenotype?. We hypothesized that there is synergistic efficacy of Avelumab (anti-PD-L1) combined with HFRT in pts with secondarily trans- formed IDH-mutant GBMs. Safety-lead-in results will be presented.
METHOD(S): This is a phase II, open-label, single-arm, multicenter study of Avelumab with HFRT in adults with transformed IDH-mutant GBM who previously received RT and TMZ and/or PCV. All pts received Avelumab 10 mg/kg IV followed at Day 8 by HFRT (25 Gy in 5 daily 5-Gy fractions) and then Avelumab 10 mg/kg IV every 2 weeks. A 3 + 3 design was used for a 6-patient safety-lead-in cohort. Adverse events were recorded according to CTCAE.
RESULT(S): Six pts (F=4, M=2) with a median age= 45.5 yrs (range 31.5-54.4 yrs) were enrolled in the safety-lead-in cohort. No DLT was observed. Grade >= 3 AEs included increased cerebral edema (3 pts), hyponatremia (1 pt) and worsening hemiparesis (3 pts). Grade <= 2 AEs included nausea, hypothyroidism, lymphopenia, thrombocytopenia, transaminase elevation, and fever/chills. Median follow-up time was 8.9 mo. Best treatment response was SD in 1 patient. At time of last follow-up all pts have discontinued treatment for PD. Median PFS was 4.2 mo (range 1.4-5.7). Median OS was 10.1 (range 6.8-21+) mo. 4 pts (67%) died, 2 pts remain alive in follow-up at 6.9 and 21.6 months after treatment initiation. The study was closed after the safety lead-in completed enrollment due to slow accrual.
CONCLUSION(S): Avelumab combined with HFRT was tolerable without dose-limiting toxicity in this safety-lead-in cohort of adult patients with transformed IDH-mutant GBM. Further studies are necessary to determine efficacy of this treatment regimen

BACKGROUND: While most hemangioblastomas (~70%) are sporadic and occur predominantly in the cerebellum, they may present as well as familial form associated with von Hippel-Lindau (VHL) syndrome, an autosomal dominant disorder caused by germline mutations of the VHL gene that trigger nuclear translocation of hypoxia-inducible factor (HIF)- 1alpha and angiogenesis. Although inactivation of VHL, a tumor suppressor gene, has been observed in hemangioblastomas, the underlying pathogenic mechanisms responsible for familial and sporadic hemangioblastomas remain incompletely understood.
METHOD(S): Whole exome sequencing of cerebellar hemangioblastoma tumors and matched blood leukocytes from 24 patients, age 24-63, was performed. After preparation and amplification of barcoded libraries, exomes were captured using Kapa Biosystems methodology and paired-end sequenced on Illumina HiSeq 2500 to an average 100-fold coverage. Following read alignment to hg19 genome, tumor and germline (leukocyte) sequences were compared, and pathogenic single nucleotide variants (SNVs) identified and validated by re-sequencing followed by pathway analysis. Additionally, tumor RNA isolated using Maxwell Promega was sequenced on Illumina instrument and the expression counts determined and normalized.
RESULT(S): We found 314 pathogenic and/or highly deleterious mutations (both germline and somatic) with a median of 13 mutations per patient. Five patients had VHL syndrome (germline VHL mutation) and 4 carried somatic VHL mutations. Among the VHL tumors, 82 mutations were identified, including HNF1B, NOTCH1 and TCF7L1, suggesting a potential contribution of altered RNA metabolism based upon pathway analysis. Among all hemangioblastomas, germline growth factor receptor variants (FGFR4 p.G388R (14/23 (61%) patients), IGF1R, PDGFRA and TYK2) known to activate STAT3 signaling and induce HIF-1alpha and angiogenesis, were identified. Non-hierarchical clustering of RNA sequencing data revealed two transcriptionally-distinct subtypes of hemangioblastomas.
CONCLUSION(S): Our findings indicate that hemangioblastomas can also occur by germline mutations known to activate STAT3 signaling, which may have significant implication in genetic testing and counseling of patients with hemangioblastomas

Medical innovators and regulators have increasingly recognized the importance of working with patients to design medical therapies and clinical trials that meet the needs of specific patient populations. For diseases such as Parkinson's disease (PD), a progressive, degenerative disease with few effective treatment options, traditional randomized clinical trials with a fixed statistical threshold may not reflect patients' perspectives on the trade-off between the risk of endorsing an ineffective therapy (false positive) and the risk of rejecting an effective therapy (false negative). This collaborative project, which involved academia, industry, FDA, patient-scientists and MJFF, developed and tested methods for incorporating patient preference information as explicit means to set significance levels in clinical trial design.
Method(s): With direct input from patients with PD, we developed a patient preference survey and deployed it online through Fox Insight for 6-weeks and received 2,752 complete responses (24.4%), allowing us to analyze differences in outcome priorities among various demographic groups. We then assigned weights to the consequences of errors based on identified patient preferences, and proposed a hypothetical clinical trial design optimized to maximize the values identified by patients.
Result(s): Movement symptoms, which are common endpoints in PD clinical trials, were ranked as most important, and psychological and cognitive symptoms, which are less commonly studied, were ranked as the next most important. Differences emerged from different groups within the patient population, depending upon how the disease manifested itself. Preferences from respondents with mild PD symptoms and no prior experience with deep brains stimulation (

Oral cancer patients report severe function-induced pain; severity is greater in females. We hypothesize that a neutrophil-mediated endogenous analgesic mechanism is responsible for sex differences in nociception secondary to oral squamous cell carcinoma (SCC). Neutrophils isolated from the cancer-induced inflammatory microenvironment contain β-endorphin protein and are identified by the Ly6G⁺ immune marker. We previously demonstrated that male mice with carcinogen-induced oral SCC exhibit less nociceptive behavior and a higher concentration of neutrophils in the cancer microenvironment compared to female mice with oral SCC. Oral cancer cells secrete granulocyte colony stimulating factor (G-CSF), a growth factor that recruits neutrophils from bone marrow to the cancer microenvironment. We found that recombinant G-CSF (rG-CSF, 5 μg/mouse, intraperitoneal) significantly increased circulating Ly6G⁺ neutrophils in the blood of male and female mice within 24 h of administration. In an oral cancer supernatant mouse model, rG-CSF treatment increased cancer-recruited Ly6G⁺ neutrophil infiltration and abolished orofacial nociceptive behavior evoked in response to oral cancer supernatant in both male and female mice. Local naloxone treatment restored the cancer mediator-induced nociceptive behavior. We infer that rG-CSF-induced Ly6G⁺ neutrophils drive an endogenous analgesic mechanism. We then evaluated the efficacy of chronic rG-CSF administration to attenuate oral cancer-induced nociception using a tongue xenograft cancer model with the HSC-3 human oral cancer cell line. Saline-treated male mice with HSC-3 tumors exhibited less oral cancer-induced nociceptive behavior and had more β-endorphin protein in the cancer microenvironment than saline-treated female mice with HSC-3 tumors. Chronic rG-CSF treatment (2.5 μg/mouse, every 72 h) increased the HSC-3 recruited Ly6G⁺ neutrophils, increased β-endorphin protein content in the tongue and attenuated nociceptive behavior in female mice with HSC-3 tumors. From these data, we conclude that neutrophil-mediated endogenous opioids warrant further investigation as a potential strategy for oral cancer pain treatment.