Comparing effectiveness of image perturbation and test retest imaging in improving radiomic model reliability.

Image perturbation is a promising technique to assess radiomic feature repeatability, but whether it can achieve the same effect as test-retest imaging on model reliability is unknown. This study aimed to compare radiomic model reliability based on repeatable features determined by the two methods using four different classifiers. A 191-patient public breast cancer dataset with 71 test-retest scans was used with pre-determined 117 training and 74 testing samples. We collected apparent diffusion coefficient images and manual tumor segmentations for radiomic feature extraction. Random translations, rotations, and contour randomizations were performed on the training images, and intra-class correlation coefficient (ICC) was used to filter high repeatable features. We evaluated model reliability in both internal generalizability and robustness, which were quantified by training and testing AUC and prediction ICC. Higher testing performance was found at higher feature ICC thresholds, but it dropped significantly at ICC = 0.95 for the test-retest model. Similar optimal reliability can be achieved with testing AUC = 0.7-0.8 and prediction ICC > 0.9 at the ICC threshold of 0.9. It is recommended to include feature repeatability analysis using image perturbation in any radiomic study when test-retest is not feasible, but care should be taken when deciding the optimal feature repeatability criteria.

Real-world experience of cabozantinib in Asian patients with advanced renal cell carcinoma following treatment with VEGFR tyrosine kinase inhibitors and/or immune-checkpoint inhibitors.

Background: There is a lack of real-world data on the use of cabozantinib in Asian patients with metastatic renal cell carcinoma. Methods: We conducted a retrospective study to investigate the toxicity and efficacy of cabozantinib in this patient population who progressed on tyrosine kinase inhibitors and/or immune-checkpoint inhibitors from six oncology centres in Hong Kong. The primary endpoint was the incidence of serious adverse events (AEs) attributed to cabozantinib. Secondary safety endpoints included dose reductions and AE-led treatment terminations. Secondary effectiveness endpoints included overall survival, progression-free survival, and objective response rate. Results: A total of 24 patients were included. Half received cabozantinib as a third-line or later-line treatment, whilst 50% received prior immune-checkpoint inhibitors, primarily nivolumab. Overall, 13 (54.2%) patients reported at least one cabozantinib-related AE of grades 3-4. The most commonly reported AEs were hand-foot skin reactions (9; 37.5%) and anaemia (4; 16.7%). Fifteen (65.2%) patients required dose reductions. Three patients discontinued treatment because of AEs. The median progression-free survival and overall survival were 10.3 months and 13.2 months, respectively; 6 (25%) patients achieved partial responses, and 8 (33.3%) achieved stable disease. Conclusion: Cabozantinib was generally well tolerated and efficacious in Asian patients with metastatic renal cell carcinoma who were heavily pretreated.

Treatment Outcomes of Computer Tomography-Guided Brachytherapy in Cervical Cancer in Hong Kong: A Retrospective Review.

(1) Background: To report the long-term clinical outcomes of computer-tomography (CT)-guided brachytherapy (BT) for locally advanced cervical cancer. (2) Methods: A total of 135 patients with FIGO stage IB-IVA cervical cancer treated with definitive radiotherapy +/- chemotherapy with an IGABT boost at Queen Mary Hospital, Hong Kong, between November 2013 and December 2019 were included. Treatment included pelvic radiotherapy 40 Gy/20 Fr/4 weeks +/- chemotherapy then CT-guided BT (7 Gy × 4 Fr) and a sequential parametrial boost. The primary outcome was local control. Secondary outcomes were pelvic control, distant metastasis-free survival, overall survival (OS) and late toxicities. (3) Results: The median follow-up was 53.6 months (3.0-99.6 months). The five-year local control, pelvic control, distant metastasis-free survival and OS rates were 90.7%, 84.3%, 80.0% and 87.2%, respectively. The incidence of G3/4 long-term toxicities was 6.7%, including proctitis (2.2%), radiation cystitis (1.5%), bowel perforation (0.7%), ureteric stricture (0.7%) and vaginal stenosis and fistula (0.7%). Patients with adenocarcinomas had worse local control (HR 5.82, 95% CI 1.84-18.34, p = 0.003), pelvic control (HR 4.41, 95% CI 1.83-10.60, p = 0.001), distant metastasis-free survival (HR 2.83, 95% CI 1.17-6.84, p = 0.021) and OS (HR 4.38, 95% CI: 1.52-12.67, p = 0.003) rates. Distant metastasis-free survival was associated with HR-CTV volume ≥ 30 cm3 (HR 3.44, 95% CI 1.18-9.42, p = 0.025) and the presence of pelvic lymph node (HR 3.44, 95% CI 1.18-9.42, p = 0.025). OS was better in patients with concurrent chemotherapy (HR 4.33, 95% CI: 1.40-13.33, p = 0.011). (4) Conclusions: CT-guided BT for cervical cancer achieved excellent long-term local control and OS. Adenocarcinoma was associated with worse clinical outcomes. (4) Conclusion: CT-guided BT for cervical cancer achieved excellent long-term local control and OS. Adenocarcinoma was associated with worse clinical outcomes.

Association between MRI histogram features and treatment response in locally advanced cervical cancer treated by chemoradiotherapy.

OBJECTIVE: To examine the associations of histogram features of T2-weighted (T2W) images and apparent diffusion coefficient (ADC) with treatment response in locally advanced cervical cancer (LACC) following concurrent chemoradiotherapy (CCRT). MATERIALS AND METHODS: Fifty-eight patients who underwent a 4-week CCRT regimen with MRI prior to treatment (pre-CCRT) and after treatment (post-CCRT) were retrospectively analysed. Histogram features were calculated from volumes of interest (VOIs) from one radiologist on T2W images and ADC maps. VOIs from two radiologists were used to assess observer repeatability in delineation and feature values at both time-points with the Dice similarity coefficient (DSC) and intraclass correlation coefficient (ICC). Treatment response was defined as a 90% reduction in tumour volume. Paired Mann-Whitney U tests were used to determine if features changed significantly between examinations. Two-sample Mann-Whitney U tests were used to identify features that were significantly different between response groups. Receiver operating characteristic (ROC) analysis was done on significantly different MRI features between treatment response groups. RESULTS: Pre-CCRT delineation and feature repeatability were generally good (DSC > 0.700; ICC > 0.750). Post-CCRT repeatability was low (DSC < 0.700; ICC < 0.750), but ADC mean and percentiles retained good ICC scores. All features, except for T2WKurtosis, significantly changed between examinations. Post-CCRT ADC50 was the only feature that demonstrated both good observer variability and significant differences between treatment response groups (p = 0.036) and had an AUC of 0.701 with a cut-off of 1.357 × 10-6 mm2/s. CONCLUSION: ADC and T2W histogram features could be used to track changes in LACC tumours undergoing CCRT. Post-CCRT ADC50 was associated with treatment response with good observer repeatability. KEY POINTS: • Pre-treatment tumour delineation and histogram feature values had good observer repeatability, while these were less repeatable at post-treatment. • MRI histogram analysis could be used to track changes in the tumour as it undergoes concurrent chemoradiotherapy. • Post-treatment median ADC was associated with treatment response and had good repeatability.

Third-line systemic treatment versus best supportive care for advanced/metastatic gastric cancer: A systematic review and meta-analysis.

This review evaluated the efficacy, toxicities and quality of life of third-line systemic treatment (TLT) versus best supportive care (BSC) in metastatic gastric cancer patients after failing two lines of systemic treatment. Six studies were included, involving 890 participants (TLT: 587, BSC: 303, Asian: 679, 76.3%), median 53-61 years old, ECOG 0-1 with no major co-morbidities. Compared with BSC, TLT improved overall survival (HR 0.63; 95% CI 0.46-0.87, corresponding to an improvement in medial OS from 3.20 to 4.80 months), progression-free survival (HR 0.29; 95% CI 0.18-0.45), objective response rate (RR 5.28; 95% CI 1.00-27.83) and disease control rate (RR 4.51; 95% CI 2.64-7.71). The efficacy results favoring TLT should be interpreted with caution for the substantial heterogeneities, wide confidence intervals and selection bias. More toxicities occurred in the TLT arms. This review highlighted the paucity of QOL data. Future studies should focus more on QOL-related outcomes. PROSPERO registration: 2015 CRD42015017873.

Intravoxel incoherent motion MRI assessment of chemoradiation-induced pelvic bone marrow changes in cervical cancer and correlation with hematological toxicity.

PURPOSE: To investigate bone marrow changes after chemoradiation (CRT) using intravoxel incoherent motion magnetic resonance imaging (IVIM-MRI) and correlate imaging changes with hematological toxicity (HT) in patients with locally advanced cervical cancer. MATERIALS AND METHODS: Thirty-nine patients with newly diagnosed cervical cancer were prospectively recruited for two sequential 3.0T IVIM-MRI studies: before treatment (MRI-1) and 3-4 weeks after standardized CRT (MRI-2). The irradiated pelvic bone marrow was outlined as the regions of interest to derive the true diffusion coefficient (D) and perfusion fraction (f) based on a biexponential model. The apparent coefficient diffusion (ADC) was derived using the monoexponential model. Changes in these parameters between MRI-1 and MRI-2 were calculated as ΔD, Δf, and ΔADC. HT was defined accordingly to NCI-CTCAE (v. 4.03) of grade 3 and above. Statistical analysis was performed using Mann-Whitney U-test. RESULTS: The median age of patients was 54 years old (range 27-83 years old); 14 patients suffered from HT. Early bone marrow changes (3-4 weeks) of ΔD showed a significant difference between HT and non-HT groups (6.4 ± 19.7% vs. -6.4 ± 19.4%, respectively, P = 0.041). However, no significant changes were noted in Δf (3.7 ± 13.3% vs. 1.5 ± 12.5% respectively, P = 0. 592) and ΔADC (5.5 ± 26.3% vs. -3.3 ± 27.0% respectively, P = 0.303) between the HT and non-HT groups. Δf increased insignificantly for both groups. CONCLUSION: ΔD was the only significant parameter to differentiate early cellular environment changes in bone marrow after CRT, suggestive that ΔD was more sensitive than Δf and ΔADC to reflect the underlying microenvironment injury. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2017;46:1491-1498.

Perfusion and diffusion characteristics of cervical cancer based on intraxovel incoherent motion MR imaging-a pilot study.

OBJECTIVES: To investigate the tissue characteristics of cervical cancer based on the intravoxel incoherent motion (IVIM) model and to assess the IVIM parameters in tissue differentiation in the female pelvis. METHODS: Sixteen treatment-naïve cervical cancer and 17 age-matched healthy subjects were prospectively recruited for diffusion-weighted (b = 0-1,000 s/mm(2)) and standard pelvic MRI. Bi-exponential analysis was performed to derive the perfusion parameters f (perfusion fraction) and D* (pseudodiffusion coefficient) as well as the diffusion parameter D (true molecular diffusion coefficient) in cervical cancer (n = 16), normal cervix (n = 17), myometrium (n = 33) and leiomyoma (n = 14). Apparent diffusion coefficient (ADC) was calculated. Kruskal-Wallis test and receiver operating characteristics (ROC) curves were used. RESULTS: Cervical cancer had the lowest f (14.9 ± 2.6%) and was significantly different from normal cervix and leiomyoma (p < 0.05). The D (0.86 ± 0.16 x 10(-3) mm2/s) was lowest in cervical cancer and was significantly different from normal cervix and myometrium (p < 0.05) but not leiomyoma. No difference was observed in D*. D was consistently lower than ADC in all tissues. ROC curves indicated that f < 16.38%, D < 1.04 × 10(-3) mm(2)/s and ADC < 1.13 × 10(-3) mm(2)/s could differentiate cervical cancer from non-malignant tissues (AUC 0.773-0.908). CONCLUSIONS: Cervical cancer has low perfusion and diffusion IVIM characteristics with promising potential for tissue differentiation. KEY POINTS: • Diffusion-weighted MRI is increasingly applied in evaluation of cervical cancer. • Cervical cancer has distinctive perfusion and diffusion characteristics. • Intravoxel incoherent motion characteristics can differentiate cervical cancer from non-malignant uterine tissues.