Team-Based Surgical Scheduling for Improved Patient Access in a High-Volume, Tertiary Head and Neck Cancer Center.

BACKGROUND: Delays in care can lead to inferior survival outcomes in head and neck cancer and other cancers. In the case of malignancies for which surgery is the preferred primary treatment modality, challenges in surgical scheduling can present a major hurdle to initiating definitive therapy in a timely fashion. It is critical to maintain efficient use of operating room resources. Traditionally, surgery is scheduled with the surgeon who initially saw the patient in consultation, and timing of surgery is tightly linked to the availability and operating room block time of the individual surgeon. METHODS: Scheduling of oncologic head and neck surgery was transitioned from a surgeon-specific method to a team-based approach wherein a patient in need of oncologic head and neck surgery is scheduled with the next-available surgeon with appropriate expertise. RESULTS: Despite substantial growth of our practice, transition to a team-based scheduling approach allowed us to maintain high utilization of operating room block time. Patient and surgeon satisfaction remain high with this new system. CONCLUSIONS: A team-based surgical scheduling approach can help optimize operating room utilization and minimize delays in cancer care, potentially leading to improved oncologic outcomes.

Survival, functional, and quality of life outcomes between total glossectomy with and without total laryngectomy: A narrative review.

BACKGROUND: A total glossectomy (TG) may be required for advanced tongue tumors. TG with total laryngectomy (TGL) may be indicated in some cases with tumor extension into the larynx or high risk of aspiration. Total glossectomy with laryngeal preservation (TGLP) may preserve phonation ability relative to TGL, yet TGLP may increase the risk of aspiration. METHODS: For this narrative review, we performed a comprehensive literature search of studies relevant to TG and TGL. Clinical studies investigating survival, functional outcomes, and quality of life in following TGLP or TGL were of particular interest. RESULTS: Few studies in the literature directly compare survival, functional, and quality of life (QOL) outcomes between TGLP and TGL. TGLP is associated with intelligible speech. However, studies investigating gastrostomy tube dependence following TGLP versus TGL have generated conflicting results. CONCLUSION: Further research on functional and QOL outcomes in patients undergoing TGL or TGLP is needed.

A Protocol for Resection and Immediate Reconstruction of Pediatric Mandibles Using Microvascular Free Fibula Flaps.

PURPOSE: The use of a vascularized free fibula graft (FFF) for the reconstruction of a mandible in a child with a mandibular tumor is infrequent. The purpose of this study is to report our protocol for resection of mandibular jaw tumors and immediate reconstruction using FFF in pediatric patients. METHODS: This was a retrospective case series of children with a mandibular tumor, which was resected and immediately reconstructed with FFF. All patients were treated via the same staged protocol: 1) presurgical digital planning, 2) surgical intervention (resection and immediate reconstruction), 3) postoperative care in the pediatric intensive unit, and 4) prosthodontic dental rehabilitation. Outcomes were complications and recurrence. Medical records were reviewed to document demographic information, tumor details, surgical interventions, postoperative course, and prosthodontic rehabilitation. RESULTS: Fifteen patients (10 males, average age of 13.7 years) met inclusion criteria. Ten patients had mandibular ameloblastoma. All patients were treated by a dedicated pediatric team and followed the same protocol. The average tumor size was 4.87 × 3.22 × 2.03 cm. Most fibulas (n = 12) had one osteotomy to reestablish mandibular continuity and create appropriate contour. The most common microvascular anastomosis was with a facial artery (n = 13) and the external jugular vein (n = 9). At an average follow-up of 15.5 months, there were only 3 minor donor site complications. Eight implants were placed without complications. No tumors recurred. CONCLUSIONS: The results of this study suggest that pediatric mandibular tumors can be successfully treated using a specific protocol involving resection and immediate reconstruction using FFF with minimal complications and without recurrence.

Smoking, age, nodal disease, T stage, p16 status, and risk of distant metastases in patients with squamous cell cancer of the oropharynx.

BACKGROUND: With an expectation of excellent locoregional control, ongoing efforts to de-intensify therapy for patients with human papillomavirus-associated squamous cell oropharyngeal cancer necessitate a better understanding of the metastatic risk for patients with this disease. The objective of this study was to determine what factors affect the risk of metastases in patients with squamous cell cancers of the oropharynx. METHODS: Under a shared use agreement, 547 patients from Radiation Therapy Oncology Group 0129 and 0522 with nonmetastatic oropharyngeal squamous cell cancers who had a known p16 status and smoking status were analyzed to assess the association of clinical features with the development of distant metastases. The analyzed factors included the p16 status, sex, T stage, N stage, age, and smoking history. RESULTS: A multivariate analysis of 547 patients with a median follow-up of 4.8 years revealed that an age ≥ 50 years (hazard ratio [HR], 3.28; P = .003), smoking for more than 0 pack-years (HR, 3.09; P < .001), N3 disease (HR, 2.64; P < .001), T4 disease (HR, 1.63; P = .030), and a negative p16 status (HR, 1.60; P = .044) were all factors associated with an increased risk of distant disease. CONCLUSIONS: Age, smoking, N3 disease, T4 disease, and a negative p16 status were associated with the development of distant metastases in patients with squamous cell cancers of the oropharynx treated definitively with concurrent chemoradiation.

Is There a Role for PET/CT Parameters to Characterize Benign, Malignant, and Metastatic Parotid Tumors?

OBJECTIVE: Assessment of benign and malignant lesions of the parotid gland, including metastatic lesions, is challenging with current imaging methods. Fluorine-18 FDG PET/CT is a noninvasive imaging modality that provides both anatomic and metabolic information. Semiquantitative data obtained from PET/CT, also known as PET/CT parameters, are maximum, mean, or peak standardized uptake values (SUVs); metabolic tumor volume; total lesion glycolysis; standardized added metabolic activity; and normalized standardized added metabolic activity. Our aim was to determine whether FDG PET/CT parameters can differentiate benign, malignant, and metastatic parotid tumors. MATERIALS AND METHODS: Thirty-four patients with parotid neoplasms underwent PET/CT before parotidectomy; maximum SUV, mean SUV, peak SUV, total lesion glycolysis, metabolic tumor volume, standardized added metabolic activity, and normalized standardized added metabolic activity were calculated on a dedicated workstation. Univariate analyses were performed. A ROC analysis was used to determine the ability of PET/CT parameters to predict pathologically proven benign, malignant, and metastatic parotid gland neoplasms. RESULTS: Fourteen patients had a benign or malignant primary parotid tumor. Twenty had metastases to the parotid gland. When the specificity was set to at least 85% for each parameter to identify cut points, the corresponding sensitivities ranged from 15% to 40%. Assessment of benign versus malignant lesions of parotid tumors, as well as metastasis from squamous cell carcinoma versus other metastatic causes, revealed that none of the PET/CT parameters has enough power to differentiate among these groups. CONCLUSION: PET/CT parameters, including total lesion glycolysis, metabolic tumor volume, standardized added metabolic activity, and normalized standardized added metabolic activity, are not able to differentiate benign from malignant parotid tumors, primary parotid tumors from metastasis, or metastasis from squamous cell carcinoma and nonsquamous cell carcinoma metastasis.

A model and nomogram to predict tumor site origin for squamous cell cancer confined to cervical lymph nodes.

BACKGROUND: The current study was conducted to develop a multifactorial statistical model to predict the specific head and neck (H&N) tumor site origin in cases of squamous cell carcinoma confined to the cervical lymph nodes ("unknown primaries"). METHODS: The Surveillance, Epidemiology, and End Results (SEER) database was analyzed for patients with an H&N tumor site who were diagnosed between 2004 and 2011. The SEER patients were identified according to their H&N primary tumor site and clinically positive cervical lymph node levels at the time of presentation. The SEER patient data set was randomly divided into 2 data sets for the purposes of internal split-sample validation. The effects of cervical lymph node levels, age, race, and sex on H&N primary tumor site were examined using univariate and multivariate analyses. Multivariate logistic regression models and an associated set of nomograms were developed based on relevant factors to provide probabilities of tumor site origin. RESULTS: Analysis of the SEER database identified 20,011 patients with H&N disease with both site-level and lymph node-level data. Sex, race, age, and lymph node levels were associated with primary H&N tumor site (nasopharynx, hypopharynx, oropharynx, and larynx) in the multivariate models. Internal validation techniques affirmed the accuracy of these models on separate data. CONCLUSIONS: The incorporation of epidemiologic and lymph node data into a predictive model has the potential to provide valuable guidance to clinicians in the treatment of patients with squamous cell carcinoma confined to the cervical lymph nodes.

Radiologic findings of occult nodal metastasis during clinically-N0 salvage total laryngectomy.

INTRODUCTION: Occult nodal disease (OND) during clinically-N0 salvage total laryngectomy (TL) can be detected with the Neck-Imaging-Reporting-and-Data-Systems (NI-RADS). However, some patients will still have OND revealed on final pathology. METHODS: A retrospective study on all patients who had OND during salvage TL with elective neck dissection (END) between 2009 and 2021 was performed. Repeat CT and PET scan interpretation was performed to evaluate their preoperative imaging for suspicious features. RESULTS: Among 81 salvage TL patients undergoing END, 12 (16%) had OND and a total of 26 occult nodes were identified. On pathology, the average node length [SD] was 0.6 cm [0.3]. On CT, 31% (8 of 26) had rounded morphology. On PET, most had SUVmax below blood pool. One patient scored NI-RADS 2; the rest scored 1. CONCLUSIONS: On re-review of preoperative imaging, occult nodes were subtle and challenging to identify. Despite no clear impact on survival, performing an END may provide prognostic information.

CD28-CD57+ T cells from head and neck cancer patients produce high levels of cytotoxic granules and type II interferon but are not senescent.

T lymphocytes expressing CD57 and lacking costimulatory receptors CD27/CD28 have been reported to accumulate with aging, chronic infection, and cancer. These cells are described as senescent, with inability to proliferate but enhanced cytolytic and cytokine-producing capacity. However, robust functional studies on these cells taken directly from cancer patients are lacking. We isolated these T cells and their CD27/28+ counterparts from blood and tumor samples of 50 patients with previously untreated head and neck cancer. Functional studies confirmed that these cells have enhanced ability to degranulate and produce IFN-γ. They also retain the ability to proliferate, thus are not senescent. These data suggest that CD27/28-CD57+ CD8+ T cells are a subset of highly differentiated, CD45RA+ effector memory (TEMRA) cells with retained proliferative capacity. Patients with > 34% of these cells among CD8+ T cells in the blood had a higher rate of locoregional disease relapse, suggesting these cells may have prognostic significance.

Reconstruction for Salvage Laryngectomy With Limited Pharyngectomy.

Importance: Closure technique for optimization of postoperative and functional outcomes following salvage laryngectomy remains an area of debate among head and neck surgeons. Objective: To investigate the association of salvage laryngectomy closure technique with early postoperative and functional outcomes. Design, Setting, and Participants: This retrospective cohort study included patients from 17 academic, tertiary care centers who underwent total laryngectomy with no or limited pharyngectomy after completing a course of definitive radiotherapy or chemoradiotherapy with curative intent between January 2011 and December 2016. Patients with defects not amenable to primary closure were excluded. Data were analyzed from February 14, 2021, to January 29, 2024. Exposures: Total laryngectomy with and without limited pharyngectomy, reconstructed by primary mucosal closure (PC), regional closure (RC), or free tissue transfer (FTT). Main Outcomes and Measures: Patients were stratified on the basis of the pharyngeal closure technique. Perioperative and long-term functional outcomes were evaluated with bivariate analyses. A multivariable regression model adjusted for historical risk factors for pharyngocutaneous fistula (PCF) was used to assess risk associated with closure technique. Relative risks (RRs) with 95% CIs were determined. Results: The study included 309 patients (256 [82.8%] male; mean age, 64.7 [range, 58.0-72.0] years). Defects were reconstructed as follows: FTT (161 patients [52.1%]), RC (64 [20.7%]), and PC (84 [27.2%]). A PCF was noted in 36 of 161 patients in the FTT group (22.4%), 25 of 64 in the RC group (39.1%), and 29 of 84 in the PC group (34.5%). On multivariable analysis, patients undergoing PC or RC had a higher risk of PCF compared with those undergoing FTT (PC: RR, 2.2 [95% CI, 1.1-4.4]; RC: RR, 2.5 [95% CI, 1.3-4.8]). Undergoing FTT was associated with a clinically meaningful reduction in risk of PCF (RR, 0.6; 95% CI, 0.4-0.9; number needed to treat, 7). Subgroup analysis comparing inset techniques for the RC group showed a higher risk of PCF associated with PC (RR, 1.8; 95% CI, 1.1-3.0) and predominately pectoralis myofascial flap with onlay technique (RR, 1.9; 95% CI, 1.2-3.2), but there was no association of pectoralis myocutaneous flap with cutaneous paddle interposition with PCF (RR, 1.2; 95% CI, 0.5-2.8) compared with FTT with cutaneous inset. There were no clinically significant differences in functional outcomes between the groups. Conclusion and Relevance: In this study of patients with limited pharyngeal defects, interpositional fasciocutaneous closure technique was associated with reduced risk of PCF in the salvage setting, which is most commonly achieved by FTT in academic practices. Closure technique was not associated with functional outcomes at 1 and 2 years postoperatively.

Assessing survival outcomes of patients with oral tongue squamous cell carcinoma: Focus on age, sex, and stage.

BACKGROUND: The purpose of this study was to provide further insights into whether age and/or sex are associated with prognosis in oral tongue squamous cell carcinoma. METHODS: This was a retrospective cohort study utilizing hospital registry data from 2006 to 2016 obtained from the National Cancer Database. Identified patients were divided into various cohorts based on age, sex, and staging. A descriptive analysis was performed using chi-square tests and overall survival rates were estimated using Kaplan-Meier method. RESULTS: A total of 17 642 patients were included in the study. The 5-year overall survival rates were 82.0% (95% CI: 79.8%-84.0%) in younger patients versus 67.5% (95% CI: 66.7%-68.3%, p-value <0.0001) older patients. The median overall survival for females was 143.4 months (95% CI: 133.2-NA) versus 129.8 (95% CI: 125.4-138.7, p-value <0.0001) in males. CONCLUSIONS: Our analysis suggests that younger age and female sex are both predictors of improved survival in oral tongue squamous cell carcinoma.

Free Flap Fat Volume is Not Associated With Recurrence or Wound Complications in Oral Cancer.

Objective: Adipose stem cells (ASCs) have been shown in many preclinical studies to be potent suppressors of the immune system. Prior studies suggest that ASCs may promote cancer progression and wound healing. However, clinical studies investigating the effects of native, or fat-grafted adipose tissue on cancer recurrence have generated mixed results. We investigated whether adipose content in reconstructive free flaps for oral squamous cell carcinoma (OSCC) is associated with disease recurrence and/or reduction in wound complications. Study Design: Retrospective chart review. Setting: Academic medical center. Methods: We performed a review of 55 patients undergoing free flap reconstruction for OSCC over a 14-month period. Using texture analysis software, we measured the relative free flap fat volume (FFFV) in postoperative computed tomography scans and compared fat volume with patient survival, recurrence, and wound healing complications. Results: We report no difference in mean FFFV between patients with or without recurrence: 13.47 cm3 in cancer-free survivors and 17.99 cm3 in cases that recurred (p = .56). Two-year recurrence-free survival in patients with high and low FFFV was 61.0% and 59.1%, respectively (p = .917). Although only 9 patients had wound healing complications, we found no trend in the incidence of wound healing complications between patients with high versus low FFFV. Conclusion: FFFV is not associated with recurrence or wound healing in patients undergoing free flap reconstruction for OSCC, suggesting adipose content should not be of concern to the reconstructive surgeon.

Insurance Authorization and Access to Proton Therapy for Patients With Head and Neck Cancers.

PURPOSE: We evaluated our institutional experience to assess potential racial inequities in insurance coverage for proton therapy in patients with head and neck (HN) cancer. METHODS AND MATERIALS: We examined the demographics of 1519 patients with HN cancer seen in consultation at our HN multidisciplinary clinic (HN MDC) and 805 patients for whom a proton insurance authorization was sought (PAS) from January 2020 to June 2022. The prospects for proton therapy insurance authorization were prospectively noted based on each patient's ICD-10 (International Classification of Diseases, 10th Revision) diagnosis code and their specific insurance plan. Proton-unfavorable (PU) insurance were those plans whose policy describes proton beam therapy as "experimental" or "not medically necessary" for the given diagnosis. RESULTS: For patients seen in our HN MDC, Black, Indigenous, and people of color (BIPOC) were significantly more likely to have PU insurance than non-Hispanic White (NHW) patients (24.9% vs 18.4%, P = .005). In multivariable analysis including race, average income of residence ZIP code, and Medicare eligibility age, BIPOC patients had an odds ratio of 1.25 for PU insurance (P = .041). In the PAS cohort, while there was no difference in the percentage of patients receiving insurance approval for proton therapy between NHW and BIPOC populations (88% vs 88.2%, P = .80), for patients with PU insurance, the median time to determination was significantly longer (median, 15.5 days), and the median time to start any radiation of any modality was longer (46 vs 35 days, P = .08). Compared with NHW patients, the median time from consultation to start of radiation therapy was longer for BIPOC patients (37 vs 43 days, P = .01). CONCLUSIONS: BIPOC patients were significantly more likely to have insurance plans unfavorable to proton therapy coverage. These PU insurance plans were associated with a longer median time to determination, a lower approval rate for proton therapy, and a longer time to start radiation of any modality.

Survival and functional outcomes after total glossectomy with total laryngectomy: Case series from a high-volume tertiary institution.

OBJECTIVES: Total glossectomy with total laryngectomy (TGTL) is indicated for some cases of advanced oral squamous cell carcinoma. However, this procedure is rarely performed, as quality of life outcomes are often considered poor. Consequently, few studies to date have reported survival and functional outcomes in patients undergoing TGTL. Here, we present the largest cases series to date of TGTL patients and provide relevant data on survival and functional outcomes. METHODS: Patients met inclusion criteria if they underwent TGTL (concurrent or staged) indicated for head and neck squamous cell carcinoma. Patient demographics and disease characteristics, survival outcomes, functional oral intake scores, time to oral intake, gastrostomy tube dependence, and communication methods post-surgery were retrospectively extracted from the electronic medical record. RESULTS: Survival in patients undergoing TGTL was poor. Most patients in this study were eventually approved for some oral intake of restricted consistencies but remained gastrostomy tube dependent for most of their nutritional needs. Baseline oral intake was suboptimal in most patients but often re-achieved approximately 12 months following surgery. Communication methods following surgery included writing, text-to-speech, and augmentative and alternative communication devices. CONCLUSION: Our data provide new insights comparing survival and functional outcomes of patients undergoing TGTL. Additional investigation particularly on patient-perceived quality of life following TGTL is needed to better understand the risks and benefits for patients who are candidates for TGTL.

Use of ultrasonic shears in the harvest of the free osteocutaneous fibula flap.

OBJECTIVES: Traditionally, the fibula free flap has been raised by electrocautery and sharp dissection with clipping and tying of vessels. Use of the ultrasonic scalpel has been proposed to be a faster, more hemostatic, and less traumatic method of harvest. METHODS: We performed a retrospective chart review of 58 patients who underwent fibula free flap reconstruction between 2007 and 2010. The main outcome measures were blood loss, operative time, and flap harvest time. RESULTS: Use of the ultrasonic shears was not associated with a statistically significant reduction in intraoperative blood loss, operative time, or flap harvest time. However, the flap harvest time did trend toward statistical significance (p = 0.073). Use of distribution-based effect sizes demonstrated a moderate clinically important difference in favor of the ultrasonic shears for both operative time and flap harvest time. CONCLUSIONS: The use of ultrasonic shears is comparable to traditional methods of fibula free flap harvest and can be considered an alternative method of harvest.

Surgical Resection is Justifiable for Oral T4b Squamous Cell Cancers With Masticator Space Invasion.

OBJECTIVES: To examine survival endpoints in patients with tumor (T)4b oral cavity squamous cell carcinoma (OCSCC) with pathologically proven masticator space invasion treated with primary surgery followed by adjuvant therapy. STUDY DESIGN: Retrospective review at an academic cancer center. METHODS: Twenty-five patients with T4b OCSCC with pathologic masticator space invasion were treated with primary surgery from May 2012 to December 2016. Only patients with ≥ 2 years follow-up from date of surgery were included. Sixteen patients received adjuvant chemoradiation. RESULTS: Median follow-up time was 39 months from date of surgery. Overall survival (OS), disease-specific survival (DSS), and recurrence-free survival at 24 months were 44.0%, 63.2%, and 52.6%, respectively. On univariate analyses, adjuvant chemoradiation was associated with improved OS. Advanced age and prolonged length of hospital stay was associated with worse OS. CONCLUSION: For pT4b OCSCCA involving the masticator space, primary surgical resection followed by adjuvant chemoradiation demonstrates 24-month DSS of > 50% and OS of 44%. LEVEL OF EVIDENCE: 4 Laryngoscope, 131:E466-E472, 2021.

Phase Ib Study of Chemoprevention with Green Tea Polyphenon E and Erlotinib in Patients with Advanced Premalignant Lesions (APL) of the Head and Neck.

PURPOSE: On the basis of synergistic effects between green tea polyphenon E (PPE) and EGFR-tyrosine kinase inhibitor in preclinical studies, we conducted a phase Ib study of the PPE and erlotinib combination in patients with advanced premalignant lesions (APL) of the oral cavity and larynx. PATIENTS AND METHODS: Patients were treated with a fixed dose of PPE (200 mg three times a day) and dose escalation of erlotinib (50, 75, 100 mg daily) for 6 months with tissue biopsy at baseline and 6 months. Primary endpoints were safety and toxicity; secondary endpoints were evaluation of pathologic response, cancer-free survival (CFS), overall survival (OS), and biomarker modulation. RESULTS: Among 21 enrolled patients, 19 began treatment and 17 completed 6 months of treatment with PPE and erlotinib. Main characteristics of treated patients: 15 severe dysplasia or carcinoma in situ and 17 oral cavity. Only skin rash was associated with dose-limiting toxicity and MTD. Recommended doses for phase II studies are PPE 600 mg daily plus erlotinib 100 mg daily for 6 months. Pathologic responses in 17 evaluable patients: pathologic complete response (47%) and pathologic partial response (18%). The 5-year CFS and OS were 66.3% and 93%, respectively. Among tested biomarkers, only phosphorylated ERK was correlated with response to treatment. CONCLUSIONS: Treatment with PPE and erlotinib combination was well tolerated in patients with APLs of the head and neck, and showed a high rate of pathologic response with excellent CFS. This combination deserves further investigation for the chemoprevention and/or prevention of second primary tumors in early-stage head and neck cancer.

Disparities in Postoperative Therapy for Salivary Gland Adenoid Cystic Carcinomas.

OBJECTIVES: The patterns of care for salivary gland adenoid cystic carcinomas (ACC) are unknown. We sought to assess predictors of receiving postoperative radiation and/or chemotherapy for patients with nonmetastatic, definitively resected ACC, as well as report unexpected nodal disease. METHODS: The National Cancer Data Base was queried for definitively resected nonmetastatic ACC from 2004 to 2014. Logistic regression, Kaplan-Meier, and Cox proportional-hazard models were utilized. Propensity-score matched analysis was employed to reduce confounding variables. RESULTS: A total of 3,136 patients met entry criteria: 2,252 (71.8%) received postoperative radiation, with 223 (7.4%) also receiving concurrent chemotherapy. Median follow-up was 4.87 years. In clinically lymph node negative (cN0) patients, 7.4% had pathologically positive lymph nodes (pN) + after elective neck dissection. Patients who lived closer to their treatment facility and had positive margins were more likely to receive postoperative radiation. Black patients and uninsured patients were less likely to receive radiation. Older age, male sex, advancing stage, and positive surgical margins were associated with worse overall survival (OS). With limited follow-up, receipt of radiation or chemotherapy was not associated with OS. CONCLUSION: Postoperative radiation was frequently given for resected ACC, with a minority receiving chemotherapy. Black patients and uninsured patients were less likely to receive radiation. Postoperative radiation and/or chemotherapy had no association with OS but were given in greater frequency in more advanced disease, and our series is limited by short follow-up. The disparity findings for this rare disease need to be addressed in future studies. LEVEL OF EVIDENCE: 2c Laryngoscope, 129:377-386, 2019.

Optical Biopsy of Head and Neck Cancer Using Hyperspectral Imaging and Convolutional Neural Networks.

Successful outcomes of surgical cancer resection necessitate negative, cancer-free surgical margins. Currently, tissue samples are sent to pathology for diagnostic confirmation. Hyperspectral imaging (HSI) is an emerging, non-contact optical imaging technique. A reliable optical method could serve to diagnose and biopsy specimens in real-time. Using convolutional neural networks (CNNs) as a tissue classifier, we developed a method to use HSI to perform an optical biopsy of ex-vivo surgical specimens, collected from 21 patients undergoing surgical cancer resection. Training and testing on samples from different patients, the CNN can distinguish squamous cell carcinoma (SCCa) from normal aerodigestive tract tissues with an area under the curve (AUC) of 0.82, 81% accuracy, 81% sensitivity, and 80% specificity. Additionally, normal oral tissues can be sub-classified into epithelium, muscle, and glandular mucosa using a decision tree method, with an average AUC of 0.94, 90% accuracy, 93% sensitivity, and 89% specificity. After separately training on thyroid tissue, the CNN differentiates between thyroid carcinoma and normal thyroid with an AUC of 0.95, 92% accuracy, 92% sensitivity, and 92% specificity. Moreover, the CNN can discriminate medullary thyroid carcinoma from benign multi-nodular goiter (MNG) with an AUC of 0.93, 87% accuracy, 88% sensitivity, and 85% specificity. Classical-type papillary thyroid carcinoma is differentiated from benign MNG with an AUC of 0.91, 86% accuracy, 86% sensitivity, and 86% specificity. Our preliminary results demonstrate that an HSI-based optical biopsy method using CNNs can provide multi-category diagnostic information for normal head-and-neck tissue, SCCa, and thyroid carcinomas. More patient data are needed in order to fully investigate the proposed technique to establish reliability and generalizability of the work.

Deformable Registration of Histological Cancer Margins to Gross Hyperspectral Images using Demons.

Hyperspectral imaging (HSI), a non-contact optical imaging technique, has been recently used along with machine learning technique to provide diagnostic information about ex-vivo surgical specimens for optical biopsy. The computer-aided diagnostic approach requires accurate ground truths for both training and validation. This study details a processing pipeline for registering the cancer-normal margin from a digitized histological image to the gross-level HSI of a tissue specimen. Our work incorporates an initial affine and control-point registration followed by a deformable Demons-based registration of the moving mask obtained from the histological image to the fixed mask made from the HS image. To assess registration quality, Dice similarity coefficient (DSC) measures the image overlap, visual inspection is used to evaluate the margin, and average target registration error (TRE) of needle-bored holes measures the registration error between the histologic and HSI images. Excised tissue samples from seventeen patients, 11 head and neck squamous cell carcinoma (HNSCCa) and 6 thyroid carcinoma, were registered according to the proposed method. Three registered specimens are illustrated in this paper, which demonstrate the efficacy of the registration workflow. Further work is required to apply the technique to more patient data and investigate the ability of this procedure to produce suitable gold standards for machine learning validation.