Improved ground truth annotation by multimodal image registration from 3D ultrasound to histopathology for resected tongue carcinoma.

OBJECTIVES: This study's objectives are (1) to investigate the registration accuracy from intraoperative ultrasound (US) to histopathological images, (2) to assess the agreement and correlation between measurements in registered 3D US and histopathology, and (3) to train a nnUNet model for automatic segmentation of 3D US volumes of resected tongue specimens. METHODS: Ten 3D US volumes were acquired, including the corresponding digitalized histopathological images (n = 29). Based on corresponding landmarks, the registrations between 3D US and histopathology images were calculated and evaluated using the target registration error (TRE). Tumor thickness and resection margins were measured based on three annotations: (1) manual histopathological tumor annotation (HTA), manual 3D US tumor annotation, and (2) the HTA registered in the 3D US. The agreement and correlation were computed between the measurements based on the HTA and those based on the manual US and registered HTA in US. A deep-learning model with nnUNet was trained on 151 3D US volumes. Segmentation metrics quantified the model's performance. RESULTS: The median TRE was 0.42 mm. The smallest mean difference was between registered HTA in US and histopathology with 2.16 mm (95% CI - 1.31; 5.63) and a correlation of 0.924 (p < 0.001). The nnUNet predicted the tumor with a Dice similarity coefficient of 0.621, an average surface distance of 1.15 mm, and a Hausdorff distance of 3.70 mm. CONCLUSION: Multimodal image registration enabled the HTA's registration in the US images and improved the agreement and correlation between the modalities. In the future, this could be used to annotate ground truth labels accurately.

Clinical validation of the accuracy of an intra-operative assessment tool using 3D ultrasound compared to histopathology in patients with squamous cell carcinoma of the tongue.

BACKGROUND: Histopathological analysis often shows close resection margins after surgical removal of tongue squamous cell carcinoma (TSCC). This study aimed to investigate the agreement between intraoperative 3D ultrasound (US) margin assessment and postoperative histopathology of resected TSCC. METHODS: In this study, ten patients were prospectively included. Three fiducial cannulas were inserted into the specimen. To acquire a motorized 3D US volume, the resected specimen was submerged in saline, after which images were acquired while the probe moved over the specimen. The US volumes were annotated twice: (1) automatically and (2) manually, with the automatic segmentation as initialization. After standardized histopathological processing, all hematoxylin-eosin whole slide images (WSI) were included for analysis. Corresponding US images were found based on the known WSI spacing and fiducials. Blinded observers measured the tumor thickness and the margin in the caudal, deep, and cranial directions on every slide. The anterior and posterior margin was measured per specimen. RESULTS: The mean difference in all measurements between manually segmented US and histopathology was 2.34 (SD: ±3.34) mm, and Spearman's rank correlation coefficient was 0.733 (p < 0.001). The smallest mean difference was in the tumor thickness with 0.80 (SD: ±2.44) mm and a correlation of 0.836 (p < 0.001). Limitations were observed in the caudal region, where no correlation was found. CONCLUSION: This study shows that 3D US and histopathology have a moderate to strong statistically significant correlation (r = 0.733; p < 0.001) and a mean difference between the modalities of 2.3 mm (95%CI: -4.2; 8.9). Future research should focus on patient outcomes regarding resection margins.

Prevalence of obstructive sleep apnea after treatment for advanced T-stage head and neck cancer.

PURPOSE: Treatment of head and neck cancer (HNC) may lead to obstructive sleep apnea (OSA), but conclusive results on the prevalence of OSA are lacking. The objective of this study is to investigate the prevalence of OSA in a cohort of patients treated for advanced T-stage HNC. METHODS: A cross-sectional study was conducted in two tertiary cancer care centers including patients at least 1 year after treatment with curative intent with surgery and/or (chemo)radiotherapy ((C)RT) for advanced T-staged (T3-4) cancer of the oral cavity, oropharynx, hypopharynx, or larynx. A polysomnography (PSG) was performed in all participants. OSA was defined as an apnea-hypopnea index (AHI) of 15 events/h or higher or an AHI of 5 events/h and higher with OSA related symptoms, such as sleeping problems, daytime dysfunction and/or cardiac/metabolic comorbidities collected through file review and questionnaires. RESULTS: Of the 67 participants, 48 (72%, 95% CI 59-82%) were diagnosed with OSA. Possible risk factors are male gender, higher BMI, greater neck circumference, more nicotine pack years, cardiometabolic comorbidities, use of medication with sleepiness as side effect, present tonsils, lower T-stage (T3 vs. T4 stage), higher AJCC stage and a HPV-negative tumor. CONCLUSION: In this population of advanced T-stage HNC patients, the prevalence of OSA was 72%, which is considerably higher than in the general population (2-50%). Given the high prevalence, screening of this entire subgroup for OSA may be indicated. Future studies to identify high risk factors and develop an OSA screening protocol are needed.

Implementing a deep learning model for automatic tongue tumour segmentation in ex-vivo 3-dimensional ultrasound volumes.

Three-dimensional (3D) ultrasound can assess the margins of resected tongue carcinoma during surgery. Manual segmentation (MS) is time-consuming, labour-intensive, and subject to operator variability. This study aims to investigate use of a 3D deep learning model for fast intraoperative segmentation of tongue carcinoma in 3D ultrasound volumes. Additionally, it investigates the clinical effect of automatic segmentation. A 3D No New U-Net (nnUNet) was trained on 113 manually annotated ultrasound volumes of resected tongue carcinoma. The model was implemented on a mobile workstation and clinically validated on 16 prospectively included tongue carcinoma patients. Different prediction settings were investigated. Automatic segmentations with multiple islands were adjusted by selecting the best-representing island. The final margin status (FMS) based on automatic, semi-automatic, and manual segmentation was computed and compared with the histopathological margin. The standard 3D nnUNet resulted in the best-performing automatic segmentation with a mean (SD) Dice volumetric score of 0.65 (0.30), Dice surface score of 0.73 (0.26), average surface distance of 0.44 (0.61) mm, Hausdorff distance of 6.65 (8.84) mm, and prediction time of 8 seconds. FMS based on automatic segmentation had a low correlation with histopathology (r = 0.12, p = 0.67); MS resulted in a moderate but insignificant correlation with histopathology (r = 0.4, p = 0.12, n = 16). Implementing the 3D nnUNet yielded fast, automatic segmentation of tongue carcinoma in 3D ultrasound volumes. Correlation between FMS and histopathology obtained from these segmentations was lower than the moderate correlation between MS and histopathology.

Comparison of image quality of 3D ultrasound: motorized acquisition versus freehand navigated acquisition, a phantom study.

PURPOSE: Intra-operative assessment of resection margins during oncological surgery is a field that needs improvement. Ultrasound (US) shows the potential to fulfill this need, but this imaging technique is highly operator-dependent. A 3D US image of the whole specimen may remedy the operator dependence. This study aims to compare and evaluate the image quality of 3D US between freehand acquisition (FA) and motorized acquisition (MA). METHODS: Multiple 3D US volumes of a commercial phantom were acquired in motorized and freehand fashion. FA images were collected with electromagnetic navigation. An integrated algorithm reconstructed the FA images. MA images were stacked into a 3D volume. The image quality is evaluated following the metrics: contrast resolution, axial and elevation resolution, axial and elevation distance calibration, stability, inter-operator variability, and intra-operator variability. A linear mixed model determined statistical differences between FA and MA for these metrics. RESULTS: The MA results in a statistically significant lower error of axial distance calibration (p < 0.0001) and higher stability (p < 0.0001) than FA. On the other hand, the FA has a better elevation resolution (p < 0.003) than the MA. CONCLUSION: MA results in better image quality of 3D US than the FA method based on axial distance calibration, stability, and variability. This study suggests acquiring 3D US volumes for intra-operative ex vivo margin assessment in a motorized fashion.

[Oral cavity squamous cell carcinoma also occurs in children]. / Plaveiselcelcarcinoom in mondholte komt ook voor bij kinderen.

Squamous cell carcinoma in the head and neck area is rare in children and constitutes only 2% of malignancies. Between October 2019 and December 2020, 3 young, male patients presented at the Princess Máxima Centre for Paediatric Oncology with squamous cell carcinoma of the oral mucosa. All 3 had complaints of increasing, painful swelling in the oral cavity for weeks to months prior to the diagnosis. They had no risk factors for developing an oral malignancy and blank medical histories. In all 3 cases, there was a long delay preceding the diagnosis and the diagnosis was made at an advanced stage. Although rare, primary squamous cell carcinoma of the oral cavity can also occur in children.

Personalized biomechanical tongue models based on diffusion-weighted MRI and validated using optical tracking of range of motion.

For advanced tongue cancer, the choice between surgery and organ-sparing treatment is often dependent on the expected loss of tongue functionality after treatment. Biomechanical models might assist in this choice by simulating the post-treatment function loss. However, this function loss varies between patients and should, therefore, be predicted for each patient individually. In the present study, the goal was to better predict the postoperative range of motion (ROM) of the tongue by personalizing biomechanical models using diffusion-weighted MRI and constrained spherical deconvolution reconstructions of tongue muscle architecture. Diffusion-weighted MRI scans of ten healthy volunteers were obtained to reconstruct their tongue musculature, which were subsequently registered to a previously described population average or atlas. Using the displacement fields obtained from the registration, the segmented muscle fiber tracks from the atlas were morphed back to create personalized muscle fiber tracks. Finite element models were created from the fiber tracks of the atlas and those of the individual tongues. Via inverse simulation of a protruding, downward, left and right movement, the ROM of the tongue was predicted. This prediction was compared to the ROM measured with a 3D camera. It was demonstrated that biomechanical models with personalized muscles bundles are better in approaching the measured ROM than a generic model. However, to achieve this result a correction factor was needed to compensate for the small magnitude of motion of the model. Future versions of these models may have the potential to improve the estimation of function loss after treatment for advanced tongue cancer.

The Timed Swallowing Proficiency for Eating and Drinking (SPEAD) Test: Development and Initial Validation of an Instrument to Objectify (Impaired) Swallowing Capacity in Head and Neck Cancer Patients.

Objective swallowing outcomes measure the physical swallowing function, while subjective outcomes measure swallowing perception. A test for swallowing capacity, measuring the ingestion of all consistencies is currently not available. Therefore, the Swallowing Proficiency for Eating And Drinking (SPEAD) test was developed. It entails the timed ingestion of thin liquid, thick liquid and solid. In this study, its feasibility, reliability and validity were evaluated in patients with dysphagia after treatment for head and neck cancer (HNC) and healthy participants. Thirty-eight HNC patients and forty healthy participants were enrolled in this study and performed the SPEAD test three times. Video recordings of the test were evaluated three times by one observer, and once by three additional observers, to assess test-retest, intra-rater and inter-rater reliability. Validity was assessed by calculating effect sizes for the difference between results of patients and healthy participants and by evaluating correlations with objective (e.g., videofluoroscopy and functional oral intake scale) and subjective (e.g., SWAL-QOL) swallowing outcomes. Test-retest, intra-rater and inter-rater reliability of ingestion duration was good to excellent. All hypotheses with regard to magnitude and direction of correlations were confirmed, supporting construct validity of the test. Our initial results suggest that the SPEAD test reliably measures the transport capacity of the upper digestive tract (in grams per second) and that this test can be useful to objectively evaluate and monitor the (safe) swallowing capacity in HNC patients, in both research as well as daily clinical practice.

Quantification of tongue mobility impairment using optical tracking in patients after receiving primary surgery or chemoradiation.

PURPOSE: Tongue mobility has shown to be a clinically interesting parameter on functional results after tongue cancer treatment which can be objectified by measuring the Range Of Motion (ROM). Reliable measurements of ROM would enable us to quantify the severity of functional impairments and use these for shared decision making in treatment choices, rehabilitation of speech and swallowing disturbances after treatment. METHOD: Nineteen healthy participants, eighteen post-chemotherapy patients and seventeen post-surgery patients were asked to perform standardized tongue maneuvers in front of a 3D camera system, which were subsequently tracked and corrected for head and jaw motion. Indicators, such as the left-right tongue range and the deflection angle with the horizontal axis were extracted from the tongue trajectory to serve as a quantitative measure for the impaired tongue mobility. RESULTS: The range and deflection angle showed an excellent intra- and interrater reliability (ICC 0.9) The repeatability experiment showed an average standard deviation of 2.5 mm to 3.5 mm for every movement, except the upward movement. The post-surgery patient group showed a smaller tongue range and higher deflection angle overall than the healthy participants. Post-chemoradiation patients showed less difference in tongue ROM compared with healthy participants. Only a few patients showed asymmetrical movement after treatment, which could not always be explained by T-stage or the side of treatment alone. CONCLUSION: We introduced a reliable and reproducible method for measuring the ROM and to quantify for motion impairments, that was able to show differences in tongue ROM between healthy subjects and patients after chemoradiation or surgery. Future research should focus on measuring patients with oral cancer pre- and post-treatment in combination with the collection of detailed information about the individual tongue anatomy, so that the full ROM trajectory can be used to identify changes over time and to quantify functional impairment.

An interactive surgical simulation tool to assess the consequences of a partial glossectomy on a biomechanical model of the tongue.

Oral cancer surgery has a negative influence on the quality of life (QOL). As a result of the complex physiology involved in oral functions, estimation of surgical effects on functionality remains difficult. We present a user-friendly biomechanical simulation of tongue surgery, including closure with suturing and scar formation, followed by an automated adaptation of a finite element (FE) model to the shape of the tongue. Different configurations of our FE model were evaluated and compared to a well-established FE model. We showed that the post-operative impairment as predicted by our model was qualitatively comparable to a patient case for five different tongue maneuvers.

From reactive to proactive tube feeding during chemoradiotherapy for head and neck cancer: A clinical prediction model-based approach.

OBJECTIVES: Feeding tubes are placed unnecessarily in a proportion of head and neck cancer (HNC) patients treated with chemoradiotherapy (CRT) when prophylactic tube placement protocols are used. This may have a negative impact on the risk of long-term dysphagia. Reactive tube placement protocols, on the other hand, might result in weight loss and treatment interruption. The objective of this study is to identify patients at risk for prolonged tube dependency in order to implement a personalized strategy regarding proactive tube placement. MATERIALS AND METHODS: A retrospective study was performed in a consecutive cohort of HNC patients treated with primary CRT for whom a reactive tube placement protocol was used. A prediction model was developed to predict prolonged (> 90 days) feeding tube dependency. Model performance and clinical net benefit of the model were assessed. RESULTS: Of the 336 included patients, 229 (68%) needed a feeding tube during CRT and 151 (45%) were prolonged feeding tube dependent. The prediction model includes the predictors pretreatment BMI, weight loss, Functional Oral Intake Scale and T-stage. Discriminatory ability is fair (area under the ROC-curve of 0.69) and calibration is adequate (Hosmer and Lemeshow test p = .254). The model shows net benefit over current practice for probability thresholds from 35 to 80%. CONCLUSION: The developed model can be used to select patients for proactive feeding tube placement during primary CRT for HNC. The nomogram with easily obtainable parameters is a useful tool for clinicians to support shared decision making regarding proactive tube placement.

Needs and preferences of patients with head and neck cancer in integrated care.

OBJECTIVES: Incorporation of patients' perspectives in daily practice is necessary to adapt care to users' needs. However, information on patients' needs and preferences for integrated care is lacking. The aim was to explore these needs and preferences, taking patients with head and neck cancer (HNC) as example, to adapt current integrated care to be more patient-centred. DESIGN: Semi-structured interviews were held with current and former patients and chairmen of patient associations. Relevant needs and preferences were identified and categorised using the eight-dimension Picker model of patient-centred care. SETTING: Integrated HNC in the Netherlands. PARTICIPANTS: Patients with HNC and chairmen of two Dutch HNC patient associations. MAIN OUTCOME MEASURES: Patients' needs and preferences of integrated HNC care categorised according the Picker model. RESULTS: A total of 34 themes of needs and preferences were identified, by 14 patients with HNC or their delegates, using the Picker dimensions. Themes often emerged were as follows: personalisation of health care regarding patient values; clear insight into the healthcare process at organisational level; use of personalised communication, education and information that meets patients' requirements; adequate involvement of allied health professionals for physical support; more attention to the impact of HNC and its treatment; adequate involvement of family and friends; adequate general practitioner involvement in the aftercare; and waiting time reduction. CONCLUSIONS: Monitoring the identified themes in integrated HNC care, fitting in the Picker model, will enable us to respond better to the needs and preferences of patients, and patient-centred care in oncological care can be enhanced.

Variation in head and neck cancer care in the Netherlands: A retrospective cohort evaluation of incidence, treatment and outcome.

BACKGROUND: To explore variation in numbers and treatment between hospitals that treat head and neck cancer (HNC) in the Netherlands. MATERIAL AND METHODS: Patient, tumor and treatment characteristics were collected from the Netherlands Cancer Registry, while histopathological features were obtained by linkage to the national pathology record register PALGA. Inter-hospital variation in volume, stage, treatment, pathologically confirmed loco-regional recurrence and overall survival rate was evaluated by tumor site. RESULTS: In total, 2094 newly diagnosed patients were included, ranging from 65 to 417 patients in participating hospitals treating HNC in 2008. Oral cavity cancer was mainly treated by surgery only, ranging from 46 to 82% per hospital, while the proportion of surgery with (chemo)radiotherapy ranged from 18 to 40%. Increasing age, male sex, and high stage were associated with a higher hazard of dying. In oropharynx cancer, the use of (chemo)radiotherapy varied from 31 to 82% between hospitals. We found an indication that higher volume was associated with a lower overall hazard of dying for the total group, but not by subsite. Low numbers, e.g. for salivary gland, nasopharynx, nasal cavity and paranasal sinus, did not permit all desired analyses. CONCLUSION: This study revealed significant interhospital variation in numbers and treatment of especially oropharyngeal and oral cavity cancer. This study is limited because we had to rely on data recorded in the past for a different purpose. To understand whether this variation is unwanted, future research should be based on prospectively collected data, including detailed information on recurrences, additional case-mix information and cause of death.

Salivary gland pleomorphic adenoma in the Netherlands: A nationwide observational study of primary tumor incidence, malignant transformation, recurrence, and risk factors for recurrence.

INTRODUCTION: Whereas salivary gland pleomorphic adenoma (SGPA) is the most common type of salivary gland tumor, little is known about its epidemiology because national cancer registries do not register this disease. OBJECTIVES: To establish SGPA incidence trends, rates of secondary malignant transformation and recurrence and associated factors in the Netherlands. MATERIALS AND METHODS: Data on incidence, epidemiology, secondary malignant transformation and recurrence were retrieved from the Dutch pathology registry (PALGA) for the years 1992, 1997, 2002, 2007, and 2012. Multivariate analysis was performed to discover the risk factors for recurrence. RESULTS: 3506 cases of SGPA were recorded implying an overall European standardized rate of 4.2-4.9 per 100,000 person-years. Our figures showed a female preponderance (1:1.43) with an annual 1% rise in female incidence (95% confidence interval [CI]: 0.2-1.8) and a bimodal age distribution in women (p<0.0001). The overall 20-year recurrence rate was 6.7%, and median time to first recurrence was 7years. Positive and uncertain resection margins and younger age at diagnosis were risk factors for recurrence, with odds ratios (ORs) of 4.62 (95%CI 2.84-7.51), 4.08 (95%CI 2.24-7.43), and 0.42 (95%CI 0.29-0.63) respectively. Tumor locations in the minor salivary glands had lower odds of recurrence than tumors in the parotid (OR 0.24; 95% CI: 0.07-0.77; p<0.016). Malignant transformation occurred in 0.15% of SGPAs (3.2% of recurrences). CONCLUSION: This first nationwide study clearly showed sex differences in SGPA epidemiology, possibly suggesting some underlying hormonal mechanism. Long-term recurrence risks were low, and secondary malignant transformation risks were very low.

A critical evaluation of lymph node ratio in head and neck cancer.

In head and neck squamous cell carcinoma (HNSCC), the search for better prognostic factors beyond TNM-stage is ongoing. Lymph node ratio (LNR) (positive lymph nodes/total lymph nodes) is gaining interest in view of its potential prognostic significance. All HNSCC patients at the Netherlands Cancer Institute undergoing neck dissection for lymph node metastases in the neck region between 2002 and 2012 (n = 176) were included. Based on a protocol change in specimen processing, the cohort was subdivided in two distinct consecutive periods (pre and post 2007). The prognostic value of LNR, N-stage, and number of positive lymph nodes for overall survival was assessed. The mean number of examined lymph nodes after 2007 was significantly higher (42.3) than before (35.8) (p = 0.024). The higher number concerned mostly lymph nodes in level V. The mean number of positive lymph nodes before 2007 was 3.3 vs. 3.6 after 2007 (p = 0.745). By multivariate analysis of both pre- and post-2007 cohort data, two factors remained associated with an increased hazard of dying: N2 [HR 2.1 (1.1-4.1) and 2.4 (1.0-5.8)] and >3 positive lymph nodes [HR 2.0 (1.1-3.5) and 3.1 (1.4-6.9)]. Hazard ratio for LNR >7 % was not significantly different: pre 2007 at 2.2 (1.3-3.8) and post 2007 at 2.1 (1.0-4.8, p = 0.053). In this study, changes in specimen processing influenced LNR values, but not the total number of tumor positive nodes found. Therefore, in HNSCC, the number of positive nodes seems a more reliable parameter than LNR, provided a minimum number of lymph nodes are examined.

Endocrine disorders among long-term survivors of childhood head and neck rhabdomyosarcoma.

PURPOSE: Head and neck rhabdomyosarcoma (HNRMS) survivors are at increased risk of developing pituitary dysfunction as an adverse event of radiotherapy. Our aim was to investigate the frequency and risk factors for pituitary dysfunction in these survivors. Secondly, we aimed to compare the prevalence of pituitary dysfunction between survivors treated with external beam radiation therapy (EBRT) and survivors treated with the ablative surgery, moulage technique after loading brachytherapy, and surgical reconstruction (AMORE) procedure. METHODS: Eighty HNRMS survivors treated in London (EBRT based) and Amsterdam (AMORE based: AMORE if feasible, otherwise EBRT) in the period 1990-2010 and alive ≥ 2 years post-treatment were evaluated. Survivors were evaluated in multidisciplinary late-effects clinics, with measurement of linear growth, determination of thyroid function, and growth hormone parameters. Additional data, such as baseline characteristics, anthropometrics, pubertal stage, and the results of additional laboratory investigations, were retrieved from patient charts. RESULTS: Pituitary dysfunction was diagnosed in 24 in 80 (30%) survivors, after a median follow-up time of 11 years. Median time to develop pituitary dysfunction after HNRMS diagnosis was 3.0 years. Risk factors were EBRT-based therapy (odds ratio [OR] 2.06; 95% confidence interval [CI] 1.79-2.46), parameningeal tumour site (OR 1.83; 95% CI 1.60-2.17) and embryonal RMS histology (OR 1.49; 95% CI 1.19-1.90). CONCLUSIONS: Radiotherapy used for the treatment of HNRMS confers a significant risk of the development of pituitary dysfunction. AMORE-based treatment in children with HNRMS resulted in less pituitary dysfunction than treatment with conventional EBRT. Our findings underscore the importance of routine early endocrine follow-up in this specific population.

Outcome of transoral robotic surgery for stage I-II oropharyngeal cancer.

Traditionally T1-2N0 oropharyngeal carcinoma is treated with a single treatment modality, being either radiotherapy or surgery. Currently, minimally invasive surgery, such as transoral robotic surgery (TORS), is gaining popularity. The aim of this study is to assess whether T1-2N0 oropharyngeal cancer can be safely and effectively resected with TORS, and to determine the oncologic and functional outcomes. In addition, the long-term quality-of-life outcomes are reported. Between 2007 and 2012, 18 patients with early stage oropharyngeal cancers underwent transoral resection with the da Vinci robot system in the Netherlands Cancer Institute. All surviving patients filled out the self-report assessments of quality-of-life questionnaires. Median robot-assisted operating time was 115 min (range 43-186 min), while median estimated blood loss was 5 ml (range 0-125 ml). In three cases the exposure was insufficient to obtain clear tumor margins because of tumor extension and local anatomy. Fourteen patients had clear surgical margins. Four patients received adjuvant radiotherapy. Nine patients underwent an elective unilateral neck dissection. The oropharyngeal cancer recurred in two patients. Regarding the quality of life, patients who needed postoperative radiotherapy had a worse outcome and patients treated with transoral resection only did quite well. TORS seems to be an oncologically safe surgical treatment for early stage T1-2N0 oropharyngeal cancer based on this relatively small group of patients. Selecting patients in whom sufficient surgical exposure can be obtained, should be performed with the greatest care to avoid the need for adjuvant radiotherapy. Comparing radiotherapy and TORS or CO2 laser should be the next step in finding the optimal treatment for patients with T1-2N0 oropharyngeal carcinoma.