The utility of multiparametric MRI in reducing diagnostic uncertainty for primary central nervous system lymphoma

AIM

A key limitation in the early treatment initiation in primary central nervous system lymphoma (PCNSL) is the diagnostic delay caused by lack of recognition of a lesion as a possible lymphoma, steroid initiation, and lesion involution, often resulting in inconclusive biopsy. We highlight the importance of multiparametric MRI (MPMRI), which incorporates diffusion weighted imaging (DWI), dynamic susceptibility contrast-enhanced perfusion weighted imaging (DSC-PWI) and proton magnetic resonance spectroscopy (1H-MRS) in addition to standard MRI sequences in resolving diagnostic uncertainty for PCNSL.

MATERIALS AND METHODS

We present a consecutive series of 10 patients at our centre with histology-proven PCNSL (specifically, diffuse large B-cell lymphoma of the CNS) who underwent multiparametric MRI. We retrospectively analyse the qualitative and semi-quantitative parameters and assess their radiological concordance for this diagnosis.

RESULTS

We note overall low apparent diffusion coefficient on DWI (mean ADCmin of 0.74), high percentage signal recovery on perfusion weighted imaging (mean 170%), a high choline-creatine ratio and a high-grade lipid peak on MRS giving a "twin-tower" appearance. Nine of ten patients had MRMRI findings concordant for PCNSL, defined as at least 3 of 4 parameters being consistent for PCNSL.

CONCLUSION

We propose that concordance between these imaging multiparametric modalities could be used as a radiological predictor of PCNSL, reducing diagnostic delays, providing a more accurate biopsy target, and resulting in quicker treatment initiation.

Artificial intelligence for early prediction of treatment response in glioblastoma

Aims

Treatment response assessment in glioblastoma is challenging. Patients routinely undergo conventional magnetic resonance imaging (MRI), but it has a low diagnostic accuracy for distinguishing between true progression (tPD) and pseudoprogression (psPD) in the early post-chemoradiotherapy time period due to similar imaging appearances. The aim of this study was to use artificial intelligence (AI) on imaging data, clinical characteristics and molecular information within machine learning models, to distinguish between and predict early tPD from psPD in patients with glioblastoma.

Method

The study involved retrospective analysis of patients with newly-diagnosed glioblastoma over a 3.5 year period (n=340), undergoing surgery and standard chemoradiotherapy treatment, with an increase in contrast-enhancing disease on the baseline MRI study 4-6 weeks post-chemoradiotherapy. Studies had contrast-enhanced T1-weighted imaging (CE-T1WI), T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) sequences, acquired at 1.5 Tesla with 6-months follow-up to determine the reference standard outcome. 76 patients (mean age 55 years, range 18-76 years, 39% female, 46 tPD, 30 psPD) were included. Machine learning models utilised information from clinical characteristics (age, gender, resection extent, performance status), O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status and 307 quantitative imaging features; extracted from baseline study CE-T1WI/ADC and T2WI sequences using semi-automatically segmented enhancing disease and perilesional oedema masks respectively. Feature selection was performed within bootstrapped cross-validated recursive feature elimination with a random forest algorithm and Naïve Bayes five-fold cross-validation to validate the final model.

Results

Treatment response assessment based on the standard-of-care reports by clinical neuroradiologists showed an accuracy of 33% (sensitivity/specificity 52%/3%) to distinguish between tPD and psPD from the early post-treatment MRI study at 4-6 weeks. Machine learning-based models based on clinical and molecular features alone demonstrated an AUC of 0.66 and models using radiomic features alone from the early post-treatment MRI demonstrated an AUC of 0.46-0.69 depending on the feature and mask subset. A combined clinico-radiomic model utilising top common features demonstrated an AUC of 0.80 and an accuracy of 74% (sensitivity/specificity 78%/67%). The features in the final model were age, MGMT promoter methylation status, two shape-based features from the enhancing disease mask (elongation and sphericity), three radiomic features from the enhancing disease mask on ADC (kurtosis, correlation, contrast) and one radiomic feature from the perilesional oedema mask on T2WI (dependence entropy).

Conclusion

Current standard-of-care glioblastoma treatment response assessment imaging has limitations. In this study, the use of AI through a machine learning-based approach incorporating clinical characteristics and MGMT promoter methylation status with quantitative radiomic features from standard MRI sequences at early 4-6 weeks post-treatment imaging showed the best model performance and a higher accuracy to distinguish between tPD and psPD for early prediction of glioblastoma treatment response.

Machine learning-based radiomic evaluation of treatment response prediction in glioblastoma

AIM

To investigate machine learning based models combining clinical, radiomic, and molecular information to distinguish between early true progression (tPD) and pseudoprogression (psPD) in patients with glioblastoma.

MATERIALS AND METHODS

A retrospective analysis was undertaken of 76 patients (46 tPD, 30 psPD) with early enhancing disease following chemoradiotherapy for glioblastoma. Outcome was determined on follow-up until 6 months post-chemoradiotherapy. Models comprised clinical characteristics, O⁶-methylguanine-DNA methyltransferase (MGMT) promoter methylation status, and 307 quantitative imaging features extracted from enhancing disease and perilesional oedema masks on early post-chemoradiotherapy contrast-enhanced T1-weighted imaging, T2-weighted imaging (T2WI), and apparent diffusion coefficient (ADC) maps. Feature selection was performed within bootstrapped cross-validated recursive feature elimination with a random forest algorithm. Naive Bayes five-fold cross-validation was used to validate the final model.

RESULTS

Top selected features included age, MGMT promoter methylation status, two shape-based features from the enhancing disease mask, three radiomic features from the enhancing disease mask on ADC, and one radiomic feature from the perilesional oedema mask on T2WI. The final model had an area under the receiver operating characteristics curve (AUC) of 0.80, sensitivity 78.2%, specificity 66.7%, and accuracy of 73.7%.

CONCLUSION

Incorporating a machine learning-based approach using quantitative radiomic features from standard-of-care magnetic resonance imaging (MRI), in combination with clinical characteristics and MGMT promoter methylation status has a complementary effect and improves model performance for early prediction of glioblastoma treatment response.

Multiparametric MRI: practical approach and pictorial review of a useful tool in the evaluation of brain tumours and tumour-like lesions

MRI has a vital role in the assessment of intracranial lesions. Conventional MRI has limited specificity and multiparametric MRI using diffusion-weighted imaging, perfusion-weighted imaging and magnetic resonance spectroscopy allows more accurate assessment of the tissue microenvironment. The purpose of this educational pictorial review is to demonstrate the role of multiparametric MRI for diagnosis, treatment planning and for assessing treatment response, as well as providing a practical approach for performing and interpreting multiparametric MRI in the clinical setting. A variety of cases are presented to demonstrate how multiparametric MRI can help differentiate neoplastic from non-neoplastic lesions compared to conventional MRI alone.

Hepatitis C virus infection is associated with hepatic and adipose tissue insulin resistance that improves after viral cure

BACKGROUND

Chronic hepatitis C (CHC) is associated with systemic insulin resistance, yet there are limited data on the tissue-specific contribution in vivo to this adverse metabolic phenotype, and the effect of HCV cure.

METHODS

We examined tissue-specific insulin sensitivity in a cohort study involving 13 patients with CHC compared to 12 BMI-matched healthy control subjects. All subjects underwent a two-step clamp incorporating the use of stable isotopes to measure carbohydrate and lipid flux (hepatic and global insulin sensitivity) with concomitant subcutaneous adipose tissue microdialysis and biopsy (subcutaneous adipose tissue insulin sensitivity). Investigations were repeated in seven patients with CHC following antiviral therapy with a documented sustained virological response.

RESULTS

Adipose tissue was more insulin resistant in patients with CHC compared to healthy controls, as evidence by elevated glycerol production rate and impaired insulin-mediated suppression of both circulating nonesterified fatty acids (NEFA) and adipose interstitial fluid glycerol release during the hyperinsulinaemic euglycaemic clamp. Hepatic and muscle insulin sensitivity were similar between patients with CHC and controls. Following viral eradication, hepatic insulin sensitivity improved as demonstrated by a reduction in endogenous glucose production rate. In addition, circulating NEFA decreased with sustained virological response (SVR) and insulin was more effective at suppressing adipose tissue interstitial glycerol release with a parallel increase in the expression of insulin signalling cascade genes in adipose tissue consistent with enhanced adipose tissue insulin sensitivity.

CONCLUSION

Chronic hepatitis C patients have profound subcutaneous adipose tissue insulin resistance in comparison with BMI-matched controls. For the first time, we have demonstrated that viral eradication improves global, hepatic and adipose tissue insulin sensitivity.

Evaluation of Response to Stereotactic Radiosurgery in Brain Metastases Using Multiparametric Magnetic Resonance Imaging and a Review of the Literature

AIMS

Following stereotactic radiosurgery (SRS), brain metastases initially increase in size in up to a third of cases, suggesting treatment failure. Current imaging using structural magnetic resonance imaging (MRI) cannot differentiate between tumour recurrence and SRS-induced changes, creating difficulties with patient management. Combining multiparametric MRI techniques, which assess tissue physiological and metabolic information, has shown promise in answering this clinical question.

MATERIALS AND METHODS

Multiparametric MRI techniques, including spectroscopy, diffusion and perfusion imaging, were used for the differentiation of radiation-related changes and tumour recurrence after SRS for intracranial metastases in six cases. All patients presented with enlargement of the treated lesion, an increase in perilesional brain oedema and aggravation or appearance of neurological signs and symptoms from 7 to 29 weeks after primary treatment.

RESULTS

Multiparametric imaging helped to differentiate features of tumour progression (n = 4) from radiation-related changes (n = 2). A low apparent diffusion coefficient (ADC) <1000 × 10^-6 mm²/s, high relative cerebral blood volume (rCBV) ratio > 2.1, high choline:creatine (Cho:Cr) ratio > 1.8 suggested tumour recurrence. A high ADC > 1000 × 10^-6 mm²/s, low rCBV ratio < 2.1, Cho:Cr ratio < 1.8 suggested SRS-induced radiation changes. Multiparametric MRI diagnosis was confirmed by histology or radiological and clinical follow-up.

CONCLUSION

Multiparametric MRI was helpful in the early identification of radiation-related changes and tumour recurrence and may be useful for monitoring treatment changes in intracranial neoplasms after SRS treatment.