Stroke Risk in Survivors of Head and Neck Cancer.

Importance: Survivors of head and neck cancers (HNC) have increased risk of stroke. A comprehensive report using standardized methods is warranted to characterize the risk and to inform on survivorship strategy. Objective: To determine the stroke risk in subpopulations of survivors of HNC in Singapore. Design, Setting, and Participants: This national, registry-based, cross-sectional study aimed to estimate stroke risk in subgroups of the HNC population between January 2005 and December 2020. Participants were identified from the Singapore Cancer Registry, the Singapore Stroke Registry, and the Registry of Birth and Deaths using relevant International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) codes. HNC subgroups were defined based on patient demographic, disease, and treatment factors. Data were analyzed from September 2022 to September 2023. Exposure: Diagnosis of HNC. Main Outcomes and Measures: Both ischemic and hemorrhagic stroke were studied. The age-standardized incidence rate ratio (SIRR) and age-standardized incidence rate difference (SIRD) were reported. The Singapore general population (approximately 4 million) served as the reference group for these estimations. Results: A total of 9803 survivors of HNC (median [IQR] age at diagnosis, 58 [49-68] years; 7166 [73.1%] male) were identified. The most common HNC subsites were nasopharynx (4680 individuals [47.7%]), larynx (1228 individuals [12.5%]), and tongue (1059 individuals [10.8%]). A total of 337 individuals (3.4%) developed stroke over a median (IQR) follow-up of 42.5 (15.0-94.5) months. The overall SIRR was 2.46 (95% CI, 2.21-2.74), and the overall SIRD was 4.11 (95% CI, 3.37-4.85) strokes per 1000 person-years (PY). The cumulative incidence of stroke was 3% at 5 years and 7% at 10 years after HNC diagnosis. The SIRR was highest among individuals diagnosed at younger than 40 years (SIRR, 30.55 [95% CI, 16.24-52.35]). All population subsets defined by age, sex, race and ethnicity, HNC subsites (except tongue), stage, histology, and treatment modalities had increased risk of stroke compared with the general population. The SIRR and SIRD were significantly higher among individuals who had a primary radiation treatment approach (SIRR, 3.01 [95% CI, 2.64-3.43]; SIRD, 5.12 [95% CI, 4.18-6.29] strokes per 1000 PY) compared with a primary surgery approach (SIRR, 1.64 [95% CI, 1.31-2.05]; SIRD, 1.84 [95% CI, 0.923.67] strokes per 1000 PY). Conclusions and Relevance: In this cross-sectional study of survivors of HNC, elevated stroke risks were observed across different age, subsites, and treatment modalities, underscoring the importance of early screening and intervention.

Comparison of the performances of machine learning and deep learning in improving the quality of low dose lung cancer PET images.

PURPOSE: To compare the performances of machine learning (ML) and deep learning (DL) in improving the quality of low dose (LD) lung cancer PET images and the minimum counts required. MATERIALS AND METHODS: 33 standard dose (SD) PET images, were used to simulate LD PET images at seven-count levels of 0.25, 0.5, 1, 2, 5, 7.5 and 10 million (M) counts. Image quality transfer (IQT), a ML algorithm that uses decision tree and patch-sampling was compared to two DL networks-HighResNet (HRN) and deep-boosted regression (DBR). Supervised training was performed by training the ML and DL algorithms with matched-pair SD and LD images. Image quality evaluation and clinical lesion detection tasks were performed by three readers. Bias in 53 radiomic features, including mean SUV, was evaluated for all lesions. RESULTS: ML- and DL-estimated images showed higher signal and smaller error than LD images with optimal image quality recovery achieved using LD down to 5 M counts. True positive rate and false discovery rate were fairly stable beyond 5 M counts for the detection of small and large true lesions. Readers rated average or higher ratings to images estimated from LD images of count levels above 5 M only, with higher confidence in detecting true lesions. CONCLUSION: LD images with a minimum of 5 M counts (8.72 MBq for 10 min scan or 25 MBq for 3 min scan) are required for optimal clinical use of ML and DL, with slightly better but more varied performance shown by DL.

Risk of Stroke in Nasopharyngeal Cancer Survivors: A National Registry-Based Population Cohort Study.

BACKGROUND AND OBJECTIVES: We aim to determine the risk of stroke and death within 30 days after stroke in nasopharyngeal cancer (NPC) survivors. METHODS: We conducted a population-based cohort study of patients diagnosed with NPC from January 1, 2005, to December 31, 2017. Using the cancer and stroke disease registries and the Singapore general population as the reference population, we report the age-standardized incidence rate differences (SIRDs) ratios (SIRs) and the cumulative incidence of stroke and the standardized mortality rate differences (SMRDs) and ratios (SMRs) for all causes of death within 30 days after stroke for NPC survivors. RESULTS: At a median follow-up of 48.4 months (interquartile range 19.8-92.9 months) for 3,849 patients diagnosed with NPC, 96 patients developed stroke. The overall SIRD and SIR for stroke were 3.12 (95% confidence interval [CI] 2.09-4.15) and 2.54 (95% CI 2.08-3.10), respectively. The SIRD was highest for the age group 70 to 79 years old (8.84 cases per 1,000 person-years, 95% CI 0.46-17.21), while the SIR was highest for the age group 30 to 39 years old (16.41, 95% CI 6.01-35.82). The SIRD and SIR for stage 1 disease were (6.96 cases per 1000 person-years, 95% CI 2.16-11.77) and (4.15, 95% CI 2.46-7.00), respectively. The SMRD and SMR for all cause deaths within 30 days of stroke were (3.20 cases per 100 persons, 95% CI -3.87 to 10.28) and (1.34, 95% CI 0.76-2.37), respectively. DISCUSSION: The overall risk of stroke was markedly elevated in survivors of NPC, especially in stage 1 disease, compared to the general population. The risk of death within 30 days of stroke was not significantly higher for NPC survivors. CLASSIFICATION OF EVIDENCE: This study provides Class II evidence of the increased risk of stroke in survivors of NPC compared to the general population.

Radiation oncology training in Singapore - A non-Australasian RANZCR training site.

The Radiation Oncology Department at The National Cancer Institute, Singapore (NCIS) and the Royal Australian and New Zealand College of Radiology (RANZCR) has had a well-established relationship that began as a partnership to grow a pool of local radiation oncologists to meet a nation's demand for radiotherapy services. This journey has surpassed its initial aims and now has produced a generation of radiation oncologists leading a national cancer institute. We recount the history and progress of this partnership here, as well as the unique success of its product; the only RANZCR-accredited radiation oncology training site outside of Australia and New Zealand since 2002. We outline the mutual benefits through many years of collaboration and deliberate efforts to grow the partnership. We also outline the distinctive specialist training path that our trainees take to meet both the local accreditation body as well as the RANZCR requirements.

Blockchain Personal Health Records: Systematic Review.

BACKGROUND: Blockchain technology has the potential to enable more secure, transparent, and equitable data management. In the health care domain, it has been applied most frequently to electronic health records. In addition to securely managing data, blockchain has significant advantages in distributing data access, control, and ownership to end users. Due to this attribute, among others, the use of blockchain to power personal health records (PHRs) is especially appealing. OBJECTIVE: This review aims to examine the current landscape, design choices, limitations, and future directions of blockchain-based PHRs. METHODS: Adopting the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines, a cross-disciplinary systematic review was performed in July 2020 on all eligible articles, including gray literature, from the following 8 databases: ACM, IEEE Xplore, MEDLINE, ScienceDirect, Scopus, SpringerLink, Web of Science, and Google Scholar. Three reviewers independently performed a full-text review and data abstraction using a standardized data collection form. RESULTS: A total of 58 articles met the inclusion criteria. In the review, we found that the blockchain PHR space has matured over the past 5 years, from purely conceptual ideas initially to an increasing trend of publications describing prototypes and even implementations. Although the eventual application of blockchain in PHRs is intended for the health care industry, the majority of the articles were found in engineering or computer science publications. Among the blockchain PHRs described, permissioned blockchains and off-chain storage were the most common design choices. Although 18 articles described a tethered blockchain PHR, all of them were at the conceptual stage. CONCLUSIONS: This review revealed that although research interest in blockchain PHRs is increasing and that the space is maturing, this technology is still largely in the conceptual stage. Being the first systematic review on blockchain PHRs, this review should serve as a basis for future reviews to track the development of the space.

Validation of low-dose lung cancer PET-CT protocol and PET image improvement using machine learning.

PURPOSE: To conduct a simplified lesion-detection task of a low-dose (LD) PET-CT protocol for frequent lung screening using 30% of the effective PETCT dose and to investigate the feasibility of increasing clinical value of low-statistics scans using machine learning. METHODS: We acquired 33 SD PET images, of which 13 had actual LD (ALD) PET, and simulated LD (SLD) PET images at seven different count levels from the SD PET scans. We employed image quality transfer (IQT), a machine learning algorithm that performs patch-regression to map parameters from low-quality to high-quality images. At each count level, patches extracted from 23 pairs of SD/SLD PET images were used to train three IQT models - global linear, single tree, and random forest regressions with cubic patch sizes of 3 and 5 voxels. The models were then used to estimate SD images from LD images at each count level for 10 unseen subjects. Lesion-detection task was carried out on matched lesion-present and lesion-absent images. RESULTS: LD PET-CT protocol yielded lesion detectability with sensitivity of 0.98 and specificity of 1. Random forest algorithm with cubic patch size of 5 allowed further 11.7% reduction in the effective PETCT dose without compromising lesion detectability, but underestimated SUV by 30%. CONCLUSION: LD PET-CT protocol was validated for lesion detection using ALD PET scans. Substantial image quality improvement or additional dose reduction while preserving clinical values can be achieved using machine learning methods though SUV quantification may be biased and adjustment of our research protocol is required for clinical use.

Quality of reporting of cranial irradiation techniques in randomized controlled trials of primary brain tumors: A systematic review.

BACKGROUND: To assess the quality of reporting of cranial irradiation (CR) techniques in randomized controlled trials (RCTs) of primary brain tumors. METHODS: We searched PubMed and EMBASE for RCTs of primary brain tumors, published from January 1999 to November 2019 which included CR as one of the intervention arms. We assessed the initial RCTs report on whether they reported the prespecified ten criteria for CR technique adequately. Multivariable logistic regression was performed to determine the factors that were predictive of adequate quality of reporting. RESULTS: We found 85 eligible trial reports. There was significant variability in the quality of reporting among the included studies. Total radiotherapy (RT) dose and fractionation schedule were reported adequately in more than 90% of the included trials. The organs at risk dose constraints, treatment verification procedures and presence or absence of deviations in RT treatment planning and delivery were reported adequately in less than 30% of included trials. Twenty-three trials (27%) reported seven criteria or more adequately. Multivariable analysis showed that trials conducted by cooperative groups, published RT quality assurance results and having a low risk of bias in the methodological quality have higher odds of having adequate quality in reporting of CR technique (judged as adequate reporting in seven criteria or more). CONCLUSIONS: The quality of reporting on CR techniques in the RCTs of primary brain tumors is variable and suboptimal. Guidelines should be introduced to improve clarity and ensure consistency in the quality of reporting.

Applying the ASCO and European Society for Medical Oncology Value Frameworks to Nasopharyngeal Cancer Treatments: Is Adding Induction Chemotherapy or Adjuvant Chemotherapy to Concurrent Chemoradiotherapy Worthwhile?

PURPOSE: To determine and compare the incremental clinical benefit (ICB) and costs of induction chemotherapy (IC) when added to concurrent chemoradiotherapy (CCRT), concurrent chemotherapy (CC) when added to radiotherapy (RT), and CC plus adjuvant chemotherapy (AC) when added to RT for locally advanced nasopharyngeal cancer (LA-NPC). MATERIALS AND METHODS: We searched phase III randomized controlled trials (RCTs) that reported overall survival benefit with the use of IC, CC, and CC + AC in LA-NPC. We quantified the ICB using the ASCO and European Society for Medical Oncology (ESMO) value frameworks. We calculated the incremental drug costs in US dollars using the lowest average wholesale price reported in the Lexicomp drug database. RESULTS: We identified three RCTs on IC, three RCTs on CC, and four RCTs on CC + AC. The ICB was judged to be grade A based on the ESMO framework. The ASCO Net Health Benefit score ranged from 17.43 to 57.39. The incremental drug costs ranged from $133.46 to $626.14. There were no statistically significant differences in the mean Net Health Benefit scores (39.37 for IC v 37.61 for CC v 33.98 for CC + AC; P = .89) and costs ($383 for IC v $253 for CC v $460 for CC + AC; P = .27) between the three approaches. There was no statistically significant correlation between ICB and costs. CONCLUSION: The magnitudes of ICB and incremental drug costs of adding of IC to CCRT, CC to RT, and CC + AC to RT for LA-NPC are not significantly different.

Quality of reporting on thoracic radiotherapy technique in prospective lung cancer trials: A systematic review.

BACKGROUND: The aim of this study is to assess the quality of reporting of thoracic (T) RT technique for curative intent treatment in prospective lung cancer trials. METHODS: We searched MEDLINE for eligible trials published from 1996 to 2016. We assessed the included trials' reports on whether they reported the RT dose prescription method; RT dose-planning procedures; algorithm for tissue inhomogeneity dose corrections; organs at risk dose constraints; target volume definition, simulation and/or motion management procedures; treatment verification procedures; total RT dose; fractionation schedule; conduct of quality assurance as well as presence or absence of deviations in RT treatment planning and delivery adequately. We performed univariable and multivariable logistic regression to determine the factors that may influence the quality of reporting. RESULTS: We found 85 eligible trial reports. Target volume definition, total RT dose, and fractionation schedules were reported adequately in more than 90% of the included trials. Algorithm for tissue inhomogeneity dose corrections, simulation and verification procedures, presence or absence of deviations in RT treatment planning and delivery were reported adequately in less than 20% of the included trials. Twenty-three trials (27%) reported 7 criteria or more adequately. Both univariable and multivariable logistic regression showed that trials with RT focused research question were more likely to have adequate quality in reporting (judged as adequate reporting in 7 criteria or more) than trials with non-RT focused question (odds ratio 4.11, 95% confidence interval 1.10 to 15.43, P value = .04). CONCLUSION: There is significant variability in the quality of reporting on thoracic radiotherapy treatment in prospective lung cancer trials. Future research should focus on developing consensus guidelines to standardize the reporting of radiotherapy technique in clinical trials.

Quality of radiotherapy reporting in randomized controlled trials of prostate cancer.

BACKGROUND: Good radiotherapy reporting in clinical trials of prostate radiotherapy is important because it will allow accurate reproducibility of radiotherapy treatment and minimize treatment variations that can affect patient outcomes. The aim of our study is to assess the quality of prostate radiotherapy (RT) treatment reporting in randomized controlled trials in prostate cancer. METHODS: We searched MEDLINE for randomized trials of prostate cancer, published from 1996 to 2016 and included prostate RT as one of the intervention arms. We assessed if the investigators reported the ten criteria adequately in the trial reports: RT dose prescription method; RT dose-planning procedures; organs at risk (OAR) dose constraints; target volume definition, simulation procedures; treatment verification procedures; total RT dose; fractionation schedule; conduct of quality assurance (QA) as well as presence or absence of deviations in RT treatment planning and delivery. We performed multivariate logistic regression to determine the factors that may influence the quality of reporting. RESULTS: We found 59 eligible trials. There was significant variability in the quality of reporting. Target volume definition, total RT dose and fractionation schedule were reported adequately in 97% of included trials. OAR constraints, simulation procedures and presence or absence of deviations in RT treatment planning and delivery were reported adequately in 30% of included trials. Twenty-four trials (40%) reported seven criteria or more adequately. Multivariable logistic analysis showed that trials that published their quality assurance results and cooperative group trials were more likely to have adequate quality in reporting in at least seven criteria. CONCLUSION: There is significant variability in the quality of reporting on prostate radiotherapy treatment in randomized trials of prostate cancer. We need to have consensus guidelines to standardize the reporting of radiotherapy treatment in randomized trials.

Induction chemotherapy for locally advanced nasopharyngeal carcinoma treated with concurrent chemoradiation: A systematic review and meta-analysis.

PURPOSE: To determine if the addition of induction chemotherapy (IC) to concurrent chemoradiation (CCRT) in locally advanced nasopharyngeal carcinoma (LA-NPC) can improve survival. METHODS: We performed a meta-analysis of both randomized controlled trials (RCTs) and observational studies (OBS) to compare the effects of addition of IC to CCRT versus (vs) CCRT alone on overall survival (OS), progression free survival (PFS), distant metastasis-free survival (DMFS) and adverse events (AE) in LA-NPC. We searched MEDLINE for eligible studies comparing IC plus CCRT vs CCRT for LA-NPC from Jan 1996 to May 2017. We selected RCTs and OBS that included patients with non-metastatic, LA-NPC who received IC followed by CCRT or CCRT alone. Three reviewers independently assessed the abstracts for eligibility. We assessed the methodological quality of the included studies using the MERGE criteria. We performed the meta-analysis with random effects model. We used the GRADE approach to appraise the quality of evidence from RCTs. The primary outcome was OS; secondary outcomes included PFS, DMFS and AE. RESULTS: We found six RCTs and five OBS including 2802 patients with low to moderate risk of bias in their methodological quality. There was high quality evidence from the RCTs that IC improved PFS (HR 0.69, 95% CI 0.57-0.84, P = 0.0003, I2 = 0%) and OS (HR 0.77, 95% CI 0.60-0.98, P = 0.03, I2 = 0%) significantly and was associated with more frequent AE. The estimates of IC effects from RCTs and OBS were similar (PFS HR 0.69 vs 0.71, interaction P (IP) = 0.92; OS HR 0.77 vs 0.58, IP = 0.27). CONCLUSIONS: IC delays disease progression and improves survival significantly for LA-NPC treated with CCRT, and was associated with more toxicity. There were no divergent results between RCTs and OBS. IC followed by CCRT can be considered one of the standard treatment options for LA-NPC.

Quality of head and neck radiotherapy reporting in randomized controlled trials.

BACKGROUND: The purpose of this clinical review was to determine the quality reporting of radiotherapy (RT) in randomized controlled trials (RCTs) of head and neck cancer and its impact on reporting of the trial primary efficacy and toxicity outcomes. METHODS: We searched MEDLINE for eligible RCTs published between 1994 and 2015. We assessed the quality of RT reporting and bias in the reporting of trial outcomes using published criteria. RESULTS: We found 67 eligible trial reports. There was significant variability in the quality of RT treatment reporting among the included trials. Thirty-two trials had adequate quality RT reporting. Cooperative group trials were more likely to have adequate quality reporting (odds ratio [OR] 3.02; 95% confidence interval [CI] 1.04-9.85). The quality of RT reporting did not influence the bias in the reporting of trial outcomes. CONCLUSION: The quality of head and neck RT reporting in RCTs is variable and did not impact on bias in reporting of trial outcomes.

Feasibility study of toxicity outcomes using GEC-ESTRO contouring guidelines on CT based instead of MRI-based planning in locally advanced cervical cancer patients.

PURPOSE: To report late rectal and bladder toxicity outcomes of a CT-based image-guided brachytherapy (IGBT) technique for treatment of cervical cancer. METHODS AND MATERIALS: Between 2008 and 2014, 95 women with International Federation of Gynecology and Obstetrics stage IB to IVA cervical carcinoma treated with definitive concurrent cisplatin-based chemotherapy and external beam radiation therapy 50.4 Gy in 28 fractions followed by planned prescription dose of 7 Gy × 4 fractions of high-dose-rate IGBT was retrospectively reviewed. At each implantation, all patients had a urinary catheter in situ and received bowel enema before undergoing planning CT simulation. A high-risk clinical target volume (HRCTV) as per GEC-ESTRO guidelines and the entire cervix, rectum, and bladder was contoured on the simulation CT according to Radiation Therapy Oncology Group Gynaecology Contouring Atlas. Reported doses to HRCTV and organs at risk were recorded. Toxicities were recorded using National Cancer Institute Common Terminology Criteria for Adverse Events version 3. RESULTS: The median followup time was 29 months. The mean HRCTV equivalent dose in 2 Gy fractions (EQD2) of external beam radiation therapy combined with brachytherapy was 80 Gy (standard deviation [SD], 11), and the rectal doses to 2 cm3 (D2cc) EQD2 and bladder D2cc EQD2 were 74 Gy (SD, 6) and 79 Gy (SD, 15), respectively. Twenty-two patients (23%) had grade 2 proctitis and 10 patients (11%) had grade 3 proctitis. Four patients (4%) had grade 2 cystitis and two patients (2%) had grade 3 cystitis. No patients had ≥ grade 4 toxicity. CONCLUSIONS: Despite CT-based brachytherapy planning, reported organ at risk toxicity was still significant compared with reported MRI-based planning series. Coimplementation of interstitial IGBT using the European Study on MRI-guided Brachytherapy in Locally Advanced Cervical Cancer (EMBRACE) protocol or using intensity-modulated radiation therapy during the external beam phase treatment might help to limit these late toxicities.

Prevalence and predictors of bias in the reporting of primary efficacy and toxicity endpoints in randomized clinical trials of radiation oncology.

INTRODUCTION: To determine the prevalence and predictors of bias in reporting of primary efficacy and toxicity endpoints in randomized trials (RCTs) of radiation oncology. METHODS: We searched MEDLINE for eligible RCTs published from January 1994 to October 2014. Bias in reporting of primary efficacy endpoint was defined as reporting that treatment was beneficial based on secondary endpoints despite a statistically non-significant difference in primary endpoint. Bias in reporting of toxicity endpoint was defined as not reporting toxicity findings in the abstract, discussion or results table. Logistic regression multivariate models were used to determine predictors of biased reporting. RESULTS: We found that 13% of 323 RCTs have bias in the reporting of primary efficacy endpoint with non-cooperative group trials as a significant predictor of bias (odds ratio (OR) 2.04, 95% confidence interval (CI) 1.03-4.00, P = 0.04). Thirty-five per cent of 279 RCTs were judged to have bias in the reporting of toxicity endpoint with trials not listed in Clinicaltrials.gov as a significant predictor of bias (OR 3.23, 95% CI 1.43-7.14, P = 0.004). CONCLUSION: The prevalence of bias in reporting of primary efficacy and toxicity endpoint for radiotherapy RCTs was 13% and 35% respectively. Non-cooperative group trials were more likely to have bias in the reporting of primary efficacy endpoint. Trials not listed in Clinicaltrials.gov were more likely to have bias in the reporting of toxicity endpoint.