Prediction of outcome in anal squamous cell carcinoma using radiomic feature analysis of pre-treatment FDG PET-CT.

PURPOSE: Incidence of anal squamous cell carcinoma (ASCC) is increasing, with curative chemoradiotherapy (CRT) as the primary treatment of non-metastatic disease. A significant proportion of patients have locoregional treatment failure (LRF), but distant relapse is uncommon. Accurate prognostication of progression-free survival (PFS) would help personalisation of CRT regimens. The study aim was to evaluate novel imaging pre-treatment features, to prognosticate for PFS in ASCC. METHODS: Consecutive patients with ASCC treated with curative intent at a large tertiary referral centre who underwent pre-treatment FDG-PET/CT were included. Radiomic feature extraction was performed using LIFEx software on baseline FDG-PET/CT. Outcome data (PFS) was collated from electronic patient records. Elastic net regularisation and feature selection were used for logistic regression model generation on a randomly selected training cohort and applied to a validation cohort using TRIPOD guidelines. ROC-AUC analysis was used to compare performance of a regression model encompassing standard clinical prognostic factors (age, sex, tumour and nodal stage-model A), a radiomic feature model (model B) and a combined radiomic/clinical model (model C). RESULTS: A total of 189 patients were included in the study, with 145 in the training cohort and 44 in the validation cohort. Median follow-up was 35.1 and 37. 9 months, respectively for each cohort, with 70.3% and 68.2% reaching this time-point with PFS. GLCM entropy (a measure of randomness of distribution of co-occurring pixel grey-levels), NGLDM busyness (a measure of spatial frequency of changes in intensity between nearby voxels of different grey-level), minimum CT value (lowest HU within the lesion) and SMTV (a standardized version of MTV) were selected for inclusion in the prognostic model, alongside tumour and nodal stage. AUCs for performance of model A (clinical), B (radiomic) and C (radiomic/clinical) were 0.6355, 0.7403, 0.7412 in the training cohort and 0.6024, 0.6595, 0.7381 in the validation cohort. CONCLUSION: Radiomic features extracted from pre-treatment FDG-PET/CT in patients with ASCC may provide better PFS prognosis than conventional staging parameters. With external validation, this might be useful to help personalise CRT regimens in the future.

KRAS-mutated plasma DNA as predictor of outcome from irinotecan monotherapy in metastatic colorectal cancer.

BACKGROUND: We investigated the clinical implications of KRAS and BRAF mutations detected in both archival tumor tissue and plasma cell-free DNA in metastatic colorectal cancer patients treated with irinotecan monotherapy. METHODS: Two hundred and eleven patients receiving second-line irinotecan (350 mg m(-2) q3w) were included in two independent cohorts. Plasma was obtained from pretreatment EDTA blood-samples. Mutations were detected in archival tumour and plasma with qPCR methods. RESULTS: Mutation status in tumor did not correlate to efficacy in either cohort, whereas none of the patients with mutations detectable in plasma responded to therapy. Response rate and disease control rate in plasma KRAS wt patients were 19 and 66% compared with 0 and 37%, in patients with pKRAS mutations, (P=0.04 and 0.01). Tumor KRAS status was not associated with PFS but with OS in the validation cohort. Plasma BRAF and KRAS demonstrated a strong influence on both PFS and OS. The median OS was 13.0 mo in pKRAS wt patients and 7.8 in pKRAS-mutated, (HR=2.26, P<0.0001). PFS was 4.6 and 2.7 mo, respectively (HR=1,69, P=0.01). Multivariate analysis confirmed the independent prognostic value of pKRAS status but not KRAS tumor status. CONCLUSION: Tumor KRAS has minor clinical impact, whereas plasma KRAS status seems to hold important predictive and prognostic information.

Deep Learning for Radiotherapy Outcome Prediction Using Dose Data - A Review.

Artificial intelligence, and in particular deep learning using convolutional neural networks, has been used extensively for image classification and segmentation, including on medical images for diagnosis and prognosis prediction. Use in radiotherapy prognostic modelling is still limited, however, especially as applied to toxicity and tumour response prediction from radiation dose distributions. We review and summarise studies that applied deep learning to radiotherapy dose data, in particular studies that utilised full three-dimensional dose distributions. Ten papers have reported on deep learning models for outcome prediction utilising spatial dose information, whereas four studies used reduced dimensionality (dose volume histogram) information for prediction. Many of these studies suffer from the same issues that plagued early normal tissue complication probability modelling, including small, single-institutional patient cohorts, lack of external validation, poor data and model reporting, use of late toxicity data without taking time-to-event into account, and nearly exclusive focus on clinician-reported complications. They demonstrate, however, how radiation dose, imaging and clinical data may be technically integrated in convolutional neural networks-based models; and some studies explore how deep learning may help better understand spatial variation in radiosensitivity. In general, there are a number of issues specific to the intersection of radiotherapy outcome modelling and deep learning, for example translation of model developments into treatment plan optimisation, which will require further combined effort from the radiation oncology and artificial intelligence communities.

Image-Guided Radiotherapy for Pelvic Cancers: A Review of Current Evidence and Clinical Utilisation.

The meticulous selection and utilisation of image-guided radiotherapy (IGRT) are essential for optimal radiotherapy treatment delivery when using highly conformal treatment techniques in pelvic radiotherapy. Pelvic IGRT has several general IGRT issues to consider (such as choice of match strategy, prioritisation between multiple treatment targets and margin estimates) as well as issues specific to pelvic radiotherapy, in particular large inter-fraction organ variation. A range of interventions, including adaptive treatment strategies, have been developed to address these challenges. This review covers general considerations for the clinical implementation of pelvic IGRT in routine practice and provides an overview of current knowledge regarding pelvic inter-fraction organ motion. Published IGRT evidence for each of the major tumour sites (gynaecological, prostate, bladder, rectal and anal cancer) is summarised, as are state-of-the-art adaptive approaches. General recommendations for the implementation of an institutional pelvic IGRT strategy include. •Ensuring consistency between treatment intent and the IGRT approach utilised. •Ensuring minimum national and international IGRT guidance is followed while considering the benefit of daily volumetric IGRT. •Ensuring the appropriate allied health professionals (namely therapy radiographers/radiation therapists) lead on undertaking IGRT. •Ensuring the IGRT workflow procedure is clear and includes an escalation process for difficult set-ups. •Ensuring a robust IGRT service is in place before implementing advanced adaptive approaches.

Modern hypofractionation schedules for tangential whole breast irradiation decrease the fraction size-corrected dose to the heart.

AIMS: Hypofractionation of postoperative radiotherapy for breast cancer has been evaluated in a number of large randomised clinical trials, but concerns remain over the late cardiac toxicity. In this study, we examined the predictions of the linear quadratic model on the estimated fraction size-corrected dose to the heart for four evidence-based hypofractionation regimens. MATERIALS AND METHODS: Dose plans for 60 left-sided breast cancer patients were analysed. All patients were planned with tangential fields for whole breast irradiation. Dose distributions were corrected to the equivalent dose in 2 Gy fractions (EQD(2)) using the linear quadratic model for five different fractionation schedules (50 Gy/25 fractions and four hypofractionation regimens) and for a range of α/ß values (0-5 Gy). The mean EQD(2) to the heart ( [Formula: see text] ) and the volume receiving 40 Gy ( [Formula: see text] ), both as calculated from the EQD(2) dose distributions, were compared between schedules. RESULTS: For α/ß = 3 Gy, [Formula: see text] favours hypofractionation for 40 Gy/15 fractions, 39 Gy/13 fractions and 42.5 Gy/16 fractions, but not for 41.6 Gy/13 fractions. All of the hypofractionation schedules result in lower [Formula: see text] compared with normofractionation. These results hold as long as α/ߠ≳ 1.5 Gy. If the heart is blocked from the treatment beam, the fraction size-corrected dose is lower for the first three hypofractionation schedules, compared with normofractionation, even for α/ß = âˆ¼1 Gy. CONCLUSION: For standard tangential field whole breast irradiation, most of the examined hypofractionation schedules are estimated to spare the heart when compared with normofractionation. The dose to the heart, adjusted for fraction size using the linear quadratic model, will generally be lower after hypofractionated compared with normofractionated schedules, even for very low values of α/ß.