Clinical commissioning and introduction of an in-house artificial intelligence (AI) platform for automated head and neck intensity modulated radiation therapy (IMRT) treatment planning.

BACKGROUND AND PURPOSE: To describe the clinical commissioning of an in-house artificial intelligence (AI) treatment planning platform for head-and-neck (HN) Intensity Modulated Radiation Therapy (IMRT). MATERIALS AND METHODS: The AI planning platform has three components: (1) a graphical user interface (GUI) is built within the framework of a commercial treatment planning system (TPS). The GUI allows AI models to run remotely on a designated workstation configured with GPU acceleration. (2) A template plan is automatically prepared involving both clinical and AI considerations, which include contour evaluation, isocenter placement, and beam/collimator jaw placement. (3) A well-orchestrated suite of AI models predicts optimal fluence maps, which are imported into TPS for dose calculation followed by an optional automatic fine-tuning. Six AI models provide flexible tradeoffs in parotid sparing and Planning Target Volume (PTV)-organ-at-risk (OAR) preferences. Planners could examine the plan dose distribution and make further modifications as clinically needed. The performance of the AI plans was compared to the corresponding clinical plans. RESULTS: The average plan generation time including manual operations was 10-15  min per case, with each AI model prediction taking ∼1 s. The six AI plans form a wide range of tradeoff choices between left and right parotids and between PTV and OARs compared with corresponding clinical plans, which correctly reflected their tradeoff designs. CONCLUSION: The in-house AI IMRT treatment planning platform was developed and is available for clinical use at our institution. The process demonstrates outstanding performance and robustness of the AI platform and provides sufficient validation.

Long-term risk of carotid stenosis and cerebrovascular disease after radiation therapy for head and neck cancer.

BACKGROUND: Recipients of radiation therapy (RT) for head and neck cancer (HNC) are at significantly increased risk for carotid artery stenosis (CAS) and cerebrovascular disease (CVD). We sought to determine (1) cumulative incidences of CAS and CVD among HNC survivors after RT and (2) whether CAS is associated with a RT dose response effect. METHODS: This single-institution retrospective cohort study examined patients with nonmetastatic HNC who completed (chemo)RT from January 2000 through October 2020 and subsequently received carotid imaging surveillance ≤2 years following RT completion and, in the absence of CAS, every 3 years thereafter. Exclusion criteria included history of known CAS/CVD. Asymptomatic CAS was defined as ≥50% reduction of luminal diameter, symptomatic CAS as stroke or transient ischemic attack, and composite CAS as asymptomatic or symptomatic CAS. RESULTS: Of 628 patients undergoing curative intent RT for HNC, median follow-up was 4.8 years (interquartile range, 2.6-8.3), with 97 patients followed ≥10 years. Median age was 61 years and 69% of patients received concurrent chemotherapy and 28% were treated postoperatively. Actuarial 10-year incidences of asymptomatic, symptomatic, and composite CAS were 29.6% (95% CI, 23.9-35.5), 10.1% (95% CI, 7.0-13.9), and 27.2% (95% CI, 22.5-32.1), respectively. Multivariable Cox models significant association between asymptomatic CAS and absolute carotid artery volume receiving ≥10 Gy (per mL: hazard ratio, 1.09; 95% CI, 1.02-1.16). CONCLUSIONS: HNC survivors are at high risk for post-RT CAS. A dose response effect was observed for asymptomatic CAS at doses as low as 10 Gy. PLAIN LANGUAGE SUMMARY: Recipients of radiation therapy for head and neck cancer are at significantly increased risk for carotid artery stenosis and cerebrovascular disease. However, carotid artery screening is not routinely performed among head and neck survivors following radiation therapy. In this single-institution retrospective cohort study, patients with head and neck cancer were initially screened for carotid artery stenosis ≤2 years following radiation therapy completion, then every 3 years thereafter. The 10-year actuarial incidence of carotid artery stenosis was >25% and stroke/transient ischemic attack >10%. Multivariable analysis demonstrated significant associations between asymptomatic carotid artery stenosis and artery volumes receiving ≥10 Gy.

NCCN Guidelines® Insights: Head and Neck Cancers, Version 1.2022.

The NCCN Guidelines for Head and Neck Cancers address tumors arising in the oral cavity (including mucosal lip), pharynx, larynx, and paranasal sinuses. Occult primary cancer, salivary gland cancer, and mucosal melanoma (MM) are also addressed. The specific site of disease, stage, and pathologic findings guide treatment (eg, the appropriate surgical procedure, radiation targets, dose and fractionation of radiation, indications for systemic therapy). The NCCN Head and Neck Cancers Panel meets at least annually to review comments from reviewers within their institutions, examine relevant new data from publications and abstracts, and reevaluate and update their recommendations. These NCCN Guidelines Insights summarize the panel's most recent recommendations regarding management of HPV-positive oropharynx cancer and ongoing research in this area.

Adjuvant Radiation Therapy for Clinical Stage III Melanoma in the Modern Therapeutic Era.

BACKGROUND: Adjuvant radiation therapy (RT) can decrease lymph node basin (LNB) recurrences in patients with clinically evident melanoma lymph node (LN) metastases following lymphadenectomy, but its role in the era of modern systemic therapies (ST), immune checkpoint or BRAF/MEK inhibitors, is unclear. PATIENTS AND METHODS: Patients at four institutions who underwent lymphadenectomy (1/1/2010-12/31/2019) for clinically evident melanoma LN metastases and received neoadjuvant and/or adjuvant ST with RT, or ST alone, but met indications for RT, were identified. Comparisons were made between ST alone and ST/RT groups. The primary outcome was 3-year cumulative incidence (CI) of LNB recurrence. Secondary outcomes included 3-year incidences of in-transit/distant recurrence and survival estimates. RESULTS: Of 98 patients, 76 received ST alone and 22 received ST/RT. Median follow-up time for patients alive at last follow-up was 44.6 months. The ST/RT group had fewer inguinal node metastases (ST 36.8% versus ST/RT 9.1%; P = 0.04), and more extranodal extension (ST 50% versus ST/RT 77.3%; P = 0.02) and positive lymphadenectomy margins (ST 2.6% versus ST/RT 13.6%; P = 0.04). The 3-year CI of LNB recurrences was lower for the ST/RT group compared with the ST group (13.9% versus 25.2%), but this reduction was not statistically significant (P = 0.36). Groups did not differ significantly in in-transit/distant recurrences (P = 0.24), disease-free survival (P = 0.14), or melanoma-specific survival (P = 0.20). CONCLUSIONS: In the era of modern ST, RT may still have value in reducing LNB recurrences in melanoma with clinical LN metastases. Further research should focus on whether select patient populations derive benefit from combination therapy, and optimizing indications for RT following neoadjuvant ST.

Head and Neck Cancers, Version 2.2020, NCCN Clinical Practice Guidelines in Oncology.

Treatment is complex for patients with head and neck (H&N) cancers with specific site of disease, stage, and pathologic findings guiding treatment decision-making. Treatment planning for H&N cancers involves a multidisciplinary team of experts. This article describes supportive care recommendations in the NCCN Guidelines for Head and Neck Cancers, as well as the rationale supporting a new section on imaging recommendations for patients with H&N cancers. This article also describes updates to treatment recommendations for patients with very advanced H&N cancers and salivary gland tumors, specifically systemic therapy recommendations.

NCCN Guidelines Insights: Head and Neck Cancers, Version 1.2018.

The NCCN Guidelines for Head and Neck (H&N) Cancers provide treatment recommendations for cancers of the lip, oral cavity, pharynx, larynx, ethmoid and maxillary sinuses, and salivary glands. Recommendations are also provided for occult primary of the H&N, and separate algorithms have been developed by the panel for very advanced H&N cancers. These NCCN Guidelines Insights summarize the panel's discussion and most recent recommendations regarding evaluation and treatment of nasopharyngeal carcinoma.

Immunotherapy of head and neck cancer: Emerging clinical trials from a National Cancer Institute Head and Neck Cancer Steering Committee Planning Meeting.

Recent advances have permitted successful therapeutic targeting of the immune system in head and neck squamous cell carcinoma (HNSCC). These new immunotherapeutic targets and agents are being rapidly adopted by the oncologic community and hold considerable promise. The National Cancer Institute sponsored a Clinical Trials Planning Meeting to address the issue of how to further investigate the use of immunotherapy in patients with HNSCC. The goals of the meeting were to consider phase 2 or 3 trial designs primarily in 3 different patient populations: those with previously untreated, human papillomavirus-initiated oropharyngeal cancers; those with previously untreated, human papillomavirus-negative HNSCC; and those with recurrent/metastatic HNSCC. In addition, a separate committee was formed to develop integrative biomarkers for the clinical trials. The meeting started with an overview of key immune components and principles related to HNSCC, including immunosurveillance and immune escape. Four clinical trial concepts were developed at the meeting integrating different immunotherapies with existing standards of care. These designs were presented for implementation by the head and neck committees of the National Cancer Institute-funded National Clinical Trials Network. This article summarizes the proceedings of this Clinical Trials Planning Meeting, the purpose of which was to facilitate the rigorous development and design of randomized phase 2 and 3 immunotherapeutic trials in patients with HNSCC. Although reviews usually are published immediately after the meeting is held, this report is unique because there are now tangible clinical trial designs that have been funded and put into practice and the studies are being activated to accrual. Cancer 2017;123:1259-1271. © 2016 American Cancer Society.

NCCN Guidelines Insights: Head and Neck Cancers, Version 2.2017.

The NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Head and Neck Cancers provide treatment recommendations for cancers of the lip, oral cavity, pharynx, larynx, ethmoid and maxillary sinuses, and salivary glands. Recommendations are also provided for occult primary of the head and neck (H&N), and separate algorithms have been developed by the panel for very advanced H&N cancers. These NCCN Guidelines Insights summarize the panel's discussion and most recent recommendations regarding the increase in human papillomavirus-associated oropharyngeal cancer and the availability of immunotherapy agents for treatment of patients with recurrent or metastatic H&N cancer.

Head and Neck Cancers, Version 1.2015.

These NCCN Guidelines Insights focus on recent updates to the 2015 NCCN Guidelines for Head and Neck (H&N) Cancers. These Insights describe the different types of particle therapy that may be used to treat H&N cancers, in contrast to traditional radiation therapy (RT) with photons (x-ray). Research is ongoing regarding the different types of particle therapy, including protons and carbon ions, with the goals of reducing the long-term side effects from RT and improving the therapeutic index. For the 2015 update, the NCCN H&N Cancers Panel agreed to delete recommendations for neutron therapy for salivary gland cancers, because of its limited availability, which has decreased over the past 2 decades; the small number of patients in the United States who currently receive this treatment; and concerns that the toxicity of neutron therapy may offset potential disease control advantages.

Head and neck cancers, Version 2.2014. Clinical practice guidelines in oncology.

This selection from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Head and Neck Cancers focuses on glottic laryngeal cancer, which is the most common type of laryngeal cancer and has an excellent cure rate. The lymphatic drainage of the glottis is sparse, and early stage primaries rarely spread to regional nodes. Because hoarseness is an early symptom, most glottic laryngeal cancer is early stage at diagnosis. Updates to these guidelines for 2014 include revisions to "Principles of Radiation Therapy" for each site and "Principles of Surgery," and the addition of a new section on "Principles of Dental Evaluation and Management."

Diffusion-weighted imaging for head and neck squamous cell carcinoma: quantifying repeatability to understand early treatment-induced change.

OBJECTIVE: The purpose of this study was to define baseline variability of apparent diffusion coefficient (ADC) on diffusion-weighted MR imaging (DWI) in patients with head and neck squamous cell carcinoma (HNSCC) and to compare it with early treatment-induced ADC change. SUBJECTS AND METHODS: Patients with American Joint Committee on Cancer stages III and IV HNSCC were imaged with two baseline DWI examinations 1 week apart and a third DWI examination during the 2nd week of curative-intent chemoradiation therapy. Mean ADC was measured in the primary tumor and largest lymph node for each patient on the three DWI scans. Mean baseline percentage differences (%∆ADC) were compared with intratreatment change. The repeatability coefficient for baseline %∆ADC was calculated and compared with intratreatment %∆ADC. Repeatability was also assessed with Bland-Altman plots and the intraclass correlation coefficient (ICC). RESULTS: Sixteen patients underwent double baseline imaging, with 14 also undergoing intratreatment imaging. Baseline nodal disease ADC could be measured in 16 patients, but ADC in primary tumors could only be measured in five patients. The nodal mean (SD) baseline %∆ADC was 8% (± 7%), which was significantly different compared with intratreatment changes of 32% (± 31%) (p = 0.01). Baseline ICC was 0.86 for nodal disease and 0.99 for primary tumor (excellent correlation). The calculated repeatability coefficient for baseline nodal ADC was 15%. No patients had decreases in intratreatment ADC of more than 15%. CONCLUSION: Baseline ADC variability for HNSCC is less than intratreatment ADC change for nodal disease. Assessment of response should consider intrinsic baseline variability.

Clinical outcomes of oropharyngeal squamous cell carcinoma stratified by human papillomavirus subtype: A systematic review and meta-analysis.

PURPOSE: We aim to determine if there is a survival difference between patients with oropharyngeal squamous cell carcinoma (OPSCC) associated with human papillomavirus (HPV) 16 versus HPV-non16 subtypes. PATIENT AND METHODS: Databases were queried for full length, peer-reviewed, English language, articles published between 01/01/1980 and 06/08/2022. Studies reporting clinical outcomes of OPSCC associated with HPV16 and HPV-non16 subtypes with at least 10 patients were included. Primary outcome was the overall survival (OS) of patients with HPV16- versus HPV-non16-associated OPSCC. Secondary outcomes were recurrence-free survival (RFS) and pooled rate of p16 positivity by immunohistochemistry (IHC). RESULTS: A total of 9 studies met inclusion criteria and included 1,310 patients with HPV16 and 219 with HPV-non16 subtypes of OPSCC. The prevalence of HPV-non16 was 14.3 %. The pooled 5-year OS rates for patients with HPV16 and HPV-non16 were 83.4 %(95 % CI 77.8-89.0 %) and 69.3 %(95 % CI 58.5-80.1 %), respectively. OS at 5 years was significantly worse for HPV-non16 subtype, compared to HPV16 (log odds ratio [OR] -0.54, p = 0.008). There was a trend towards worse 5-year RFS with HPV-non16 compared to HPV16 (log OR -0.55, p = 0.063). Patients with HPV-non16 disease were less likely to be p16 positive by IHC (log OR -0.91, p = 0.02). CONCLUSION: Patients with HPV-non16OPSCC may experience worse OS and were less likely to be p16 positive compared to patients with HPV16 disease. While future prospective validation is warranted, routine assessment of both p16 IHC and HPV subtype could be considered prior to pursuing treatment de-escalation for HPV-associated OPSCC.

Prognostic value of different discretization parameters in 18fluorodeoxyglucose positron emission tomography radiomics of oropharyngeal squamous cell carcinoma.

Purpose: We aim to interrogate the role of positron emission tomography (PET) image discretization parameters on the prognostic value of radiomic features in patients with oropharyngeal cancer. Approach: A prospective clinical trial (NCT01908504) enrolled patients with oropharyngeal squamous cell carcinoma (N=69; mixed HPV status) undergoing definitive radiotherapy and evaluated intra-treatment 18fluorodeoxyglucose PET as a potential imaging biomarker of early metabolic response. The primary tumor volume was manually segmented by a radiation oncologist on PET/CT images acquired two weeks into treatment (20 Gy). From this, 54 radiomic texture features were extracted. Two image discretization techniques-fixed bin number (FBN) and fixed bin size (FBS)-were considered to evaluate systematic changes in the bin number ({32, 64, 128, 256} gray levels) and bin size ({0.10, 0.15, 0.22, 0.25} bin-widths). For each discretization-specific radiomic feature space, an LASSO-regularized logistic regression model was independently trained to predict residual and/or recurrent disease. The model training was based on Monte Carlo cross-validation with a 20% testing hold-out, 50 permutations, and minor-class up-sampling to account for imbalanced outcomes data. Performance differences among the discretization-specific models were quantified via receiver operating characteristic curve analysis. A final parameter-optimized logistic regression model was developed by incorporating different settings parameterizations into the same model. Results: FBN outperformed FBS in predicting residual and/or recurrent disease. The four FBN models achieved AUC values of 0.63, 0.61, 0.65, and 0.62 for 32, 64, 128, and 256 gray levels, respectively. By contrast, the average AUC of the four FBS models was 0.53. The parameter-optimized model, comprising features joint entropy (FBN = 64) and information measure correlation 1 (FBN = 128), achieved an AUC of 0.70. Kaplan-Meier analyses identified these features to be associated with disease-free survival (p=0.0158 and p=0.0180, respectively; log-rank test). Conclusions: Our findings suggest that the prognostic value of individual radiomic features may depend on feature-specific discretization parameter settings.

Radiomics on spatial-temporal manifolds via Fokker-Planck dynamics.

BACKGROUND: Delta radiomics is a high-throughput computational technique used to describe quantitative changes in serial, time-series imaging by considering the relative change in radiomic features of images extracted at two distinct time points. Recent work has demonstrated a lack of prognostic signal of radiomic features extracted using this technique. We hypothesize that this lack of signal is due to the fundamental assumptions made when extracting features via delta radiomics, and that other methods should be investigated. PURPOSE: The purpose of this work was to show a proof-of-concept of a new radiomics paradigm for sparse, time-series imaging data, where features are extracted from a spatial-temporal manifold modeling the time evolution between images, and to assess the prognostic value on patients with oropharyngeal cancer (OPC). METHODS: To accomplish this, we developed an algorithm to mathematically describe the relationship between two images acquired at time t = 0 $t = 0$ and t > 0 $t > 0$ . These images serve as boundary conditions of a partial differential equation describing the transition from one image to the other. To solve this equation, we propagate the position and momentum of each voxel according to Fokker-Planck dynamics (i.e., a technique common in statistical mechanics). This transformation is driven by an underlying potential force uniquely determined by the equilibrium image. The solution generates a spatial-temporal manifold (3 spatial dimensions + time) from which we define dynamic radiomic features. First, our approach was numerically verified by stochastically sampling dynamic Gaussian processes of monotonically decreasing noise. The transformation from high to low noise was compared between our Fokker-Planck estimation and simulated ground-truth. To demonstrate feasibility and clinical impact, we applied our approach to 18F-FDG-PET images to estimate early metabolic response of patients (n = 57) undergoing definitive (chemo)radiation for OPC. Images were acquired pre-treatment and 2-weeks intra-treatment (after 20 Gy). Dynamic radiomic features capturing changes in texture and morphology were then extracted. Patients were partitioned into two groups based on similar dynamic radiomic feature expression via k-means clustering and compared by Kaplan-Meier analyses with log-rank tests (p < 0.05). These results were compared to conventional delta radiomics to test the added value of our approach. RESULTS: Numerical results confirmed our technique can recover image noise characteristics given sparse input data as boundary conditions. Our technique was able to model tumor shrinkage and metabolic response. While no delta radiomics features proved prognostic, Kaplan-Meier analyses identified nine significant dynamic radiomic features. The most significant feature was Gray-Level-Size-Zone-Matrix gray-level variance (p = 0.011), which demonstrated prognostic improvement over its corresponding delta radiomic feature (p = 0.722). CONCLUSIONS: We developed, verified, and demonstrated the prognostic value of a novel, physics-based radiomics approach over conventional delta radiomics via data assimilation of quantitative imaging and differential equations.

Head and neck cancers, version 2.2013. Featured updates to the NCCN guidelines.

These NCCN Guidelines Insights focus on nutrition and supportive care for patients with head and neck cancers. This topic was a recent addition to the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Head and Neck Cancers. The NCCN Guidelines Insights focus on major updates to the NCCN Guidelines and discuss the new updates in greater detail. The complete version of the NCCN Guidelines for Head and Neck Cancers is available on the NCCN Web site (NCCN.org).

Low contralateral neck recurrence risk with ipsilateral neck radiotherapy in N2b tonsillar squamous cell carcinoma.

OBJECTIVES: To characterize factors including nodal burden, pre-treatment imaging, and other patient factors which may influence the role of ipsilateral neck radiotherapy (IRT) in tonsillar squamous cell carcinoma (SCC) with multiple involved ipsilateral nodes. METHODS: Patients with cT1-2N0-2bM0 (AJCC 7th edition) tonsillar SCC treated with definitive radiation therapy (RT) at Duke University Medical Center from 1/1/1990-10/1/2019 were identified. Patient, tumor, and treatment characteristics were compared between those that received bilateral neck RT (BRT) versus IRT. Recurrence-free survival (RFS) was estimated with Kaplan-Meier method. A subset analysis of patients with N2b disease was performed. Patterns of recurrence were analyzed. RESULTS: 120 patients with cT1-2N0-2b tonsillar SCC were identified, including 71 with N2b disease (BRT: n = 30; IRT: n = 41). Median follow-up was 80 months (range: 7-209). No N2b patients who received IRT had > 1 cm of soft palate/base of tongue extension. N2b patients treated with IRT had a median of 3 (range 2-9) involved lymph nodes, with median largest nodal dimension of 2.8 cm (range 1.3-4.8 cm). 93 % of N2b patients who received IRT had staging by PET/CT, and 100 % received IMRT. For N2b patients treated with IRT, there were no contralateral neck recurrences, and 10 year RFS was 95 % (95 % CI 82 %-98 %). CONCLUSIONS: For patients treated with IRT for well-lateralized N2b tonsillar SCC, we observed high rates of local control with no observed contralateral neck recurrence. These data suggest that BRT is not universally necessary for patients with multiple involved ipsilateral nodes, particularly in the setting of baseline staging with PET/CT.

The Affordable Care Act and suicide incidence among adults with cancer.

BACKGROUND: Patients with cancer are at an increased suicide risk, and socioeconomic deprivation may further exacerbate that risk. The Affordable Care Act (ACA) expanded insurance coverage options for low-income individuals and mandated coverage of mental health care. Our objective was to quantify associations of the ACA with suicide incidence among patients with cancer. METHODS: We identified US patients with cancer aged 18-74 years diagnosed with cancer from 2011 to 2016 from the Surveillance, Epidemiology, and End Results database. The primary outcome was the 1-year incidence of suicide based on cumulative incidence analyses. Difference-in-differences (DID) analyses compared changes in suicide incidence from 2011-2013 (pre-ACA) to 2014-2016 (post-ACA) in Medicaid expansion relative to non-expansion states. We conducted falsification tests with 65-74-year-old patients with cancer, who are Medicare-eligible and not expected to benefit from ACA provisions. RESULTS: We identified 1,263,717 patients with cancer, 812 of whom died by suicide. In DID analyses, there was no change in suicide incidence after 2014 in Medicaid expansion vs. non-expansion states for nonelderly (18-64 years) patients with cancer (p = .41), but there was a decrease in suicide incidence among young adults (18-39 years) (- 64.36 per 100,000, 95% CI = - 125.96 to - 2.76, p = .041). There were no ACA-associated changes in suicide incidence among 65-74-year-old patients with cancer. CONCLUSIONS: We found an ACA-associated decrease in the incidence of suicide for some nonelderly patients with cancer, particularly young adults in Medicaid expansion vs. non-expansion states. Expanding access to health care may decrease the risk of suicide among cancer survivors.

Controversies in the locoregional management of head and neck cancer.

Head and neck cancer (HNC) is a heterogeneous combination of various sites and types of disease. This manuscript elaborates on 3 important and current issues: the emerging role of human papilloma virus (HPV) in oropharyngeal cancer (OPC), current considerations in systemic therapy for advanced disease, and evolving treatment of the neck. Exogenous carcinogens, most notably tobacco, have classically been implicated in the development of HNC. A large increase in the incidence of OPC has occurred in the past few decades, predominantly in nontobacco users, and is caused by HPV. This disease is unique in many respects and presents an opportunity for novel therapeutic approaches. Because the prognosis for HPV-related HNC is better, regardless of whether surgery or radiation is used as the primary therapy, the reduction of treatment-related morbidity has assumed increasing importance and provides unique opportunities and challenges for de-escalation of therapies. Radiotherapy (RT) and concurrent cisplatin is the most commonly used nonsurgical platform for locally advanced disease. New data suggest that viable alternatives exist to the typical 3 cycles of bolus high-dose cisplatin. The role of RT and concurrent taxanes remains less understood. Similarly, the value of integrating epidermal growth factor inhibition and concurrent chemoradiation is under continuing investigation. The use of PET scanning is changing the traditional use of adjuvant neck dissection after RT or chemoradiation. Recent data support the use of surgery in the presence of a positive posttreatment PET, and observation in the setting of a negative posttreatment scan.

Intrinsic radiomic expression patterns after 20 Gy demonstrate early metabolic response of oropharyngeal cancers.

PURPOSE: This study investigated the prognostic potential of intra-treatment PET radiomics data in patients undergoing definitive (chemo) radiation therapy for oropharyngeal cancer (OPC) on a prospective clinical trial. We hypothesized that the radiomic expression of OPC tumors after 20 Gy is associated with recurrence-free survival (RFS). MATERIALS AND METHODS: Sixty-four patients undergoing definitive (chemo)radiation for OPC were prospectively enrolled on an IRB-approved study. Investigational 18 F-FDG-PET/CT images were acquired prior to treatment and 2 weeks (20 Gy) into a seven-week course of therapy. Fifty-five quantitative radiomic features were extracted from the primary tumor as potential biomarkers of early metabolic response. An unsupervised data clustering algorithm was used to partition patients into clusters based only on their radiomic expression. Clustering results were naïvely compared to residual disease and/or subsequent recurrence and used to derive Kaplan-Meier estimators of RFS. To test whether radiomic expression provides prognostic value beyond conventional clinical features associated with head and neck cancer, multivariable Cox proportional hazards modeling was used to adjust radiomic clusters for T and N stage, HPV status, and change in tumor volume. RESULTS: While pre-treatment radiomics were not prognostic, intra-treatment radiomic expression was intrinsically associated with both residual/recurrent disease (P = 0.0256, χ 2 test) and RFS (HR = 7.53, 95% CI = 2.54-22.3; P = 0.0201). On univariate Cox analysis, radiomic cluster was associated with RFS (unadjusted HR = 2.70; 95% CI = 1.26-5.76; P = 0.0104) and maintained significance after adjustment for T, N staging, HPV status, and change in tumor volume after 20 Gy (adjusted HR = 2.69; 95% CI = 1.03-7.04; P = 0.0442). The particular radiomic characteristics associated with outcomes suggest that metabolic spatial heterogeneity after 20 Gy portends complete and durable therapeutic response. This finding is independent of baseline metabolic imaging characteristics and clinical features of head and neck cancer, thus providing prognostic advantages over existing approaches. CONCLUSIONS: Our data illustrate the prognostic value of intra-treatment metabolic image interrogation, which may potentially guide adaptive therapy strategies for OPC patients and serve as a blueprint for other disease sites. The quality of our study was strengthened by its prospective image acquisition protocol, homogenous patient cohort, relatively long patient follow-up times, and unsupervised clustering formalism that is less prone to hyper-parameter tuning and over-fitting compared to supervised learning.

Establishing ADC-Based Histogram and Texture Features for Early Treatment-Induced Changes in Head and Neck Squamous Cell Carcinoma.

The purpose of this study was to assess baseline variability in histogram and texture features derived from apparent diffusion coefficient (ADC) maps from diffusion-weighted MRI (DW-MRI) examinations and to identify early treatment-induced changes to these features in patients with head and neck squamous cell carcinoma (HNSCC) undergoing definitive chemoradiation. Patients with American Joint Committee on Cancer Stage III-IV (7th edition) HNSCC were prospectively enrolled on an IRB-approved study to undergo two pre-treatment baseline DW-MRI examinations, performed 1 week apart, and a third early intra-treatment DW-MRI examination during the second week of chemoradiation. Forty texture and six histogram features were derived from ADC maps. Repeatability of the features from the baseline ADC maps was assessed with the intra-class correlation coefficient (ICC). A Wilcoxon signed-rank test compared average baseline and early treatment feature changes. Data from nine patients were used for this study. Comparison of the two baseline ADC maps yielded 11 features with an ICC ≥ 0.80, indicating that these features had excellent repeatability: Run Gray-Level Non-Uniformity, Coarseness, Long Zone High Gray-Level, Variance (Histogram Feature), Cluster Shade, Long Zone, Variance (Texture Feature), Run Length Non-Uniformity, Correlation, Cluster Tendency, and ADC Median. The Wilcoxon signed-rank test resulted in four features with significantly different early treatment-induced changes compared to the baseline values: Run Gray-Level Non-Uniformity (p = 0.005), Run Length Non-Uniformity (p = 0.005), Coarseness (p = 0.006), and Variance (Histogram) (p = 0.006). The feasibility of histogram and texture analysis as a potential biomarker is dependent on the baseline variability of each metric, which disqualifies many features.