Radiosensitivity Index is Not Fit to be Used for Dose Adjustments: A Pan-Cancer Analysis.

AIMS: To explore the preclinical and latest clinical evidence of the radiation sensitivity signature termed 'radiosensitivity index' (RSI), to assess its suitability as an input into dose-adjustment algorithms. MATERIALS AND METHODS: The original preclinical test-set data from the publication where RSI was derived were collected and reanalysed by comparing the observed versus predicted survival fraction at 2 Gy (SF2). In addition, the predictive capability of RSI was also compared to random guessing. Clinical data were collected from a recently published dataset that included RSI values, overall survival outcomes, radiotherapy dose and tumour site for six cancers (glioma, triple-negative breast, endometrial, melanoma, pancreatic and lung cancer). Cox proportional hazards models were used to assess: (i) does adjusting for RSI elucidate a dose response and (ii) does an interaction between RSI and dose exist with good precision. RESULTS: Preclinically, RSI showed a negative correlation (Spearman's rho = -0.61) between observed and predicted SF2, which remained negative after removing leukaemia cell lines. Furthermore, random guesses showed better correlation to SF2 than RSI, 98% of the time on the full dataset and 80% after removing leukaemia cell lines. The preclinical data show that RSI does not explain the variance in SF2 better than random guessing. Clinically, a dose response was not seen after adjusting for RSI (hazard ratio = 1.00, 95% confidence interval 0.97-1.04; P = 0.876) and no evidence of an interaction between RSI and dose was found (P = 0.844). CONCLUSIONS: These results suggest that RSI does not explain a sufficient amount of the outcome variance to be used within dose-adjustment algorithms.

Improved survival prediction for oropharyngeal cancer beyond TNMv8.

PURPOSE: For oropharynx squamous cell carcinoma (OPSCC) this study aimed to: (i) compare 5-year overall survival (OS) stratification by AJCC/UICC TNM versions 7 (TNMv7) and 8 (TNMv8), (ii) determine whether changes to T and N stage groupings improve prognostication and (iii) develop and validate a model incorporating additional clinical characteristics to improve 5-year OS prediction. MATERIAL AND METHODS: All OPSCC treated with curative-intent at our institution between 2011 and 2017 were included. The primary endpoint was 5-year OS. Survival curves were produced for TNMv7 and TNMv8. A three-way interaction between T, N stage and p16 status was evaluated for improved prognostication. Cox proportional hazards modelling was used to derive a new predictive model. RESULTS: Of 750 OPSCC cases, 574 (77%) were p16-positive. TNMv8 was more prognostic than TNMv7 (concordance probability estimate [CPE] ±â€¯SE = 0.72 ±â€¯0.02 vs 0.53 ±â€¯0.02). For p16-positive disease, TNMv8 discriminated stages II vs I (HR 2.32, 95% CI 1.47-3.67) and III vs II (HR 1.75, 95% CI 1.13-2.72). For p16-negative disease, TNMv7 and TNMv8 demonstrated poor hazard discrimination. Different T, N stage and p16-status combinations did not improve prognostication after adjusting for other factors (CPE = 0.79 vs 0.79, p = 0.998). A model for p16-positive and p16-negative OPSCC including additional clinical characteristics improved 5-year OS prediction beyond TNMv8 (c-index 0.76 ±â€¯0.02). CONCLUSIONS: TNMv8 is superior to TNMv7 for p16-positive OPSCC, but both performed poorly for p16-negative disease. A novel model incorporating additional clinical characteristics improved 5-year OS prediction for both p16-positive and p16-negative disease.

Sacral Insufficiency Fracture Following Pelvic Radiotherapy in Gynaecological Malignancies: Development of a Predictive Model.

AIMS: To investigate the time-to-event and the evolution of sacral insufficiency fractures in gynaecological patients receiving pelvic external beam radiotherapy (EBRT) in relation to dosimetric and imaging parameters across a spectrum of radiotherapy delivery techniques, and to develop a predictive model with a clinical nomogram to identify those at risk of sacral insufficiency fracture. MATERIALS AND METHODS: Patients who received radical or adjuvant pelvic EBRT for gynaecological malignancy between 2014 and 2019 were identified. The data collected were: demographics and clinical details; radiotherapy planning data: dose, fractionation, technique (fixed-field intensity-modulated radiotherapy, adaptive arc, and non-adaptive arc), 60 Gy simultaneous integrated boost. Each plan was examined to determine the sacral dose in 5%/Gy3 increments. Follow-up magnetic resonance scans were reviewed for insufficiency fractures, defined as linear low T1-weighted signal intensity with a high short-T1 inversion recovery (STIR) signal. The site of insufficiency fracture was recreated on the planning computed tomography, the dose to insufficiency fracture contours was recorded and insufficiency fractures were determined as healed with resolution of high STIR signal. Univariable analysis was conducted of the clinical variables. The area under the receiver operator characteristic curve and odds ratio of the risk prediction model with 95% confidence interval are reported with a nomogram for use in clinical practice. RESULTS: 115 patients were identified; the median imaging follow-up was 12 months (2-47). 37.4% developed sacral insufficiency fractures; 93.0% were detected within 12 months of EBRT. At the final radiological follow-up, 83.7% of insufficiency fractures remained active. The radiotherapy delivery technique was not associated with insufficiency fracture after adjusting for patient age (P = 0.115). The location of the 60 Gy simultaneous integrated boost planning target volume did not impact upon the site of insufficiency fracture or the dose received by the insufficiency fracture sites. Age and V40Gy3 are predictors for insufficiency fracture and form the clinical risk model (receiver operator characteristic 0.72). CONCLUSIONS: Age and V40Gy3 predict sacral insufficiency fractures; future work should focus on optimising radiotherapy planning with adoption of a bone-sparing planning approach for those patients at high risk of insufficiency fracture.

Palliative Lung Radiotherapy: Higher Dose Leads to Improved Survival?

AIMS: Choosing the optimal palliative lung radiotherapy regimen is challenging. Guidance from The Royal College of Radiologists recommends treatment stratification based on performance status, but evidence suggests that higher radiotherapy doses may be associated with survival benefits. The aim of this study was to investigate the effects of fractionation regimen and additional factors on the survival of palliative lung cancer radiotherapy patients. MATERIALS AND METHODS: A retrospective univariable (n = 925) and multivariable (n = 422) survival analysis of the prognostic significance of baseline patient characteristics and treatment prescription was carried out on patients with non-small cell and small cell lung cancer treated with palliative lung radiotherapy. The covariates investigated included: gender, age, performance status, histology, comorbidities, stage, tumour location, tumour side, smoking status, pack year history, primary radiotherapy technique and fractionation scheme. The overall mortality rate at 30 and 90 days of treatment was calculated. RESULTS: Univariable analysis revealed that performance status (P < 0.001), fractionation scheme (P < 0.001), comorbidities (P = 0.02), small cell histology (P = 0.02), 'lifelong never' smoking status (P = 0.01) and gender (P = 0.06) were associated with survival. Upon multivariable analysis, only better performance status (P = 0.01) and increased dose/fractionation regimens of up to 30 Gy/10 fractions (P < 0.001) were associated with increased survival. Eighty-five (9.2%) and 316 patients (34%) died within 30 and 90 days of treatment, respectively. CONCLUSION: In this retrospective single-centre analysis of palliative lung radiotherapy, increased total dose (up to and including 30 Gy/10 fractions) was associated with better survival regardless of performance status.

Systems Pharmacology Modeling of Prostate-Specific Antigen in Patients With Prostate Cancer Treated With an Androgen Receptor Antagonist and Down-Regulator.

First-in-human (FIH) studies with AZD3514, a selective androgen receptor (AR) down-regulator, showed decreases of >30% in the prostate-specific antigen (PSA) in some patients. A modeling approach was adopted to understand these observations and define the optimum clinical use hypothesis for AZD3514 for clinical testing. Initial empirical modeling showed that only baseline PSA correlated significantly with this biological response, whereas drug concentration did not. To identify the mechanistic cause of this observation, a mechanism-based model was first developed, which described the effects of AZD3514 on AR protein and PSA mRNA levels in LNCaP cells with and without dihydrotestosterone (DHT). Second, the mechanism-based model was linked to a population pharmacokinetic (PK) model; PSA effects of clinical doses were subsequently simulated under different clinical conditions. This model was used to adjust the design of the ongoing clinical FIH study and direct the backup program.

An in silico canine cardiac midmyocardial action potential duration model as a tool for early drug safety assessment.

Cell lines expressing ion channels (IC) and the advent of plate-based electrophysiology device have enabled a molecular understanding of the action potential (AP) as a means of early QT assessment. We sought to develop an in silico AP (isAP) model that provides an assessment of the effect of a compound on the myocyte AP duration (APD) using concentration-effect curve data from a panel of five ICs (hNav1.5, hCav1.2, hKv4.3/hKChIP2.2, hKv7.1/hminK, hKv11.1). A test set of 53 compounds was selected to cover a range of selective and mixed IC modulators that were tested for their effects on optically measured APD. A threshold of >10% change in APD at 90% repolarization (APD(90)) was used to signify an effect at the top test concentration. To capture the variations observed in left ventricular midmyocardial myocyte APD data from 19 different dogs, the isAP model was calibrated to produce an ensemble of 19 model variants that could capture the shape and form of the APs and also quantitatively replicate dofetilide- and diltiazem-induced APD(90) changes. Provided with IC panel data only, the isAP model was then used, blinded, to predict APD(90) changes greater than 10%. At a simulated concentration of 30 µM and based on a criterion that six of the variants had to agree, isAP prediction was scored as showing greater than 80% predictivity of compound activity. Thus, early in drug discovery, the isAP model allows integrating separate IC data and is amenable to the throughput required for use as a virtual screen.