Differentiating pseudoprogression from true tumor progression has become a significant challenge in follow-up of diffuse infiltrating gliomas, particularly high grade, which leads to a potential ...treatment delay for patients with early glioma recurrence. In this study, we proposed to use a multiparametric MRI data as a sequence input for the convolutional neural network with the recurrent neural network based deep learning structure to discriminate between pseudoprogression and true tumor progression. In this study, 43 biopsy-proven patient data identified as diffuse infiltrating glioma patients whose disease progressed/recurred were used. The dataset consists of five original MRI sequences; pre-contrast T1-weighted, post-contrast T1-weighted, T2-weighted, FLAIR, and ADC images as well as two engineered sequences; T1post-T1pre and T2-FLAIR. Next, we used three CNN-LSTM models with a different set of sequences as input sequences to pass through CNN-LSTM layers. We performed threefold cross-validation in the training dataset and generated the boxplot, accuracy, and ROC curve, AUC from each trained model with the test dataset to evaluate models. The mean accuracy for VGG16 models ranged from 0.44 to 0.60 and the mean AUC ranged from 0.47 to 0.59. For CNN-LSTM model, the mean accuracy ranged from 0.62 to 0.75 and the mean AUC ranged from 0.64 to 0.81. The performance of the proposed CNN-LSTM with multiparametric sequence data was found to outperform the popular convolutional CNN with a single MRI sequence. In conclusion, incorporating all available MRI sequences into a sequence input for a CNN-LSTM model improved diagnostic performance for discriminating between pseudoprogression and true tumor progression.
To the Editor:
In the Case Record discussed by Batchelor et al. (July 23 issue),
1
induction chemotherapy with methotrexate, temozolomide, and rituximab in a woman with primary central nervous system ...(CNS) lymphoma resulted in a near-complete response manifested by minimal residual enhancement on magnetic resonance imaging. In a clinical trial, she then received treatment that included high-dose rituximab combined with cytarabine for stem-cell mobilization, followed by high-dose rituximab combined with chemotherapy with autologous stem-cell rescue. The stated rationale for the high-dose rituximab was to allow greater penetration of the CNS.
As an antibody, rituximab can be expected to enter the . . .
Assessment of radiation and chemotherapy efficacy for brain cancer patients is traditionally accomplished by measuring changes in tumor size several months after therapy has been administered. The ...ability to use noninvasive imaging during the early stages of fractionated therapy to determine whether a particular treatment will be effective would provide an opportunity to optimize individual patient management and avoid unnecessary systemic toxicity, expense, and treatment delays. We investigated whether changes in the Brownian motion of water within tumor tissue as quantified by using diffusion MRI could be used as a biomarker for early prediction of treatment response in brain cancer patients. Twenty brain tumor patients were examined by standard and diffusion MRI before initiation of treatment. Additional images were acquired 3 weeks after initiation of chemo- and/or radiotherapy. Images were coregistered to pretreatment scans, and changes in tumor water diffusion values were calculated and displayed as a functional diffusion map (fDM) for correlation with clinical response. Of the 20 patients imaged during the course of therapy, 6 were classified as having a partial response, 6 as stable disease, and 8 as progressive disease. The fDMs were found to predict patient response at 3 weeks from the start of treatment, revealing that early changes in tumor diffusion values could be used as a prognostic indicator of subsequent volumetric tumor response. Overall, fDM analysis provided an early biomarker for predicting treatment response in brain tumor patients.
Use of conventional magnetic resonance imaging (MRI) for target definition may expose glioblastomas (GB) to inadequate radiation dose coverage of the nonenhanced hypercellular subvolume. This study ...aimed to develop a technique to identify the hypercellular components of GB by using high b-value diffusion-weighted imaging (DWI) and to investigate its relationship with the prescribed 95% isodose volume (PDV) and progression-free survival (PFS).
Twenty-one patients with GB underwent chemoradiation therapy post-resection and biopsy. Radiation therapy (RT) treatment planning was based upon conventional MRI. Pre-RT DWIs were acquired in 3 orthogonal directions with b-values of 0, 1000, and 3000 s/mm(2). Hypercellularity volume (HCV) was defined on the high b-value (3000 s/mm(2)) DWI by a threshold method. Nonenhanced signified regions not covered by the Gd-enhanced gross tumor volume (GTV-Gd) on T1-weighted images. The PDV was used to evaluate spatial coverage of the HCV by the dose plan. Association between HCV and PFS or other clinical covariates were assessed using univariate proportional hazards regression models.
HCVs and nonenhanced HCVs varied from 0.58 to 67 cm(3) (median: 9.8 cm(3)) and 0.15 to 60 cm(3) (median: 2.5 cm(3)), respectively. Fourteen patients had incomplete dose coverage of the HCV, 6 of whom had >1 cm(3) HCV missed by the 95% PDV (range: 1.01-25.4 cm(3)). Of the 15 patients who progressed, 5 progressed earlier, within 6 months post-RT, and 10 patients afterward. Pre-RT HCVs within recurrent GTVs-Gd were 78% (range: 65%-89%) for the 5 earliest progressions but lower, 53% (range: 0%-85%), for the later progressions. HCV and nonenhanced HCV were significant negative prognostic indicators for PFS (P<.002 and P<.01, respectively). The hypercellularity subvolume not covered by the 95% PDV was a significant negative predictor for PFS (P<.05).
High b-value DWI identifies the hypercellular components of GB and could aid in RT target volume definition. Future studies will allow us to investigate the role of high b-value DWI in identifying radiation boost volumes and diagnosing progression.
To determine the maximum-tolerated dose (MTD) of radiation (RT) with concurrent temozolomide in patients with newly diagnosed glioblastoma (GBM), to estimate their progression-free (PFS) and overall ...survival (OS), and to assess the role of (11)C methionine PET (MET-PET) imaging in predicting recurrence.
Intensity-modulated RT (IMRT) doses of 66 to 81 Gy, assigned to patients by the time-to-event continual reassessment method, were delivered over 6 weeks with concurrent daily temozolomide (75 mg/m(2)) followed by adjuvant cyclic temozolomide (200 mg/m(2) d1-5 q28d ×6 cycles). Treatment was based on gadolinium-enhanced MRI. Pretreatment MET-PET scans were obtained for correlation with eventual sites of failure.
A total of 38 patients were analyzed with a median follow-up of 54 months for patients who remain alive. Late CNS grade ≥III toxicity was observed at 78 (2 of 7 patients) and 81 Gy (1 of 9 patients). None of 22 patients receiving 75 or less Gy developed RT necrosis. Median OS and PFS were 20.1 (14.0-32.5) and 9.0 (6.0-11.7) months, respectively. Twenty-two of 32 patients with pretreatment MET-PET uptake showed uptake beyond the contrast-enhanced MRI. Patients whose treatment did not include the region of increased MET-PET uptake showed an increased risk of noncentral failure (P < 0.001).
Patients with GBM can safely receive standard temozolomide with 75 Gy in 30 fractions, delivered using IMRT. The median OS of 20.1 months is promising. Furthermore, MET-PET appears to predict regions of high risk of recurrence not defined by MRI, suggesting that further improvements may be possible by targeting metabolically active regions.
OBJECTIVE Survival rates and prognostic factors for supratentorial hemispheric ependymomas have not been determined. The authors therefore designed a retrospective study to determine progression-free ...survival (PFS), overall survival (OS), and prognostic factors for hemispheric ependymomas. METHODS The study population consisted of 8 patients from our institution and 101 patients from the literature with disaggregated survival information (n = 109). Patient age, sex, tumor side, tumor location, extent of resection (EOR), tumor grade, postoperative chemotherapy, radiation, time to recurrence, and survival were recorded. Kaplan-Meier survival analyses and Cox proportional hazard models were completed to determine survival rates and prognostic factors. RESULTS Anaplastic histology/WHO Grade III tumors were identified in 62% of cases and correlated with older age. Three-, 5-, and 10-year PFS rates were 57%, 51%, and 42%, respectively. Three-, 5-, and 10-year OS rates were 77%, 71%, and 58%, respectively. EOR and tumor grade were identified on both Kaplan-Meier log-rank testing and univariate Cox proportional hazard models as prognostic for PFS and OS. Both EOR and tumor grade remained prognostic on multivariate analysis. Subtotal resection (STR) predicted a worse PFS (hazard ratio HR 4.764, p = 0.001) and OS (HR 4.216, p = 0.008). Subgroup survival analysis of patients with STR demonstrated a 5- and 10-year OS of 28% and 0%, respectively. WHO Grade III tumors also had worse PFS (HR 10.2, p = 0.004) and OS (HR 9.1, p = 0.035). Patients with WHO Grade III tumors demonstrated 5- and 10-year OS of 61% and 46%, respectively. Postoperative radiation was not prognostic for PFS or OS. CONCLUSIONS A high incidence of anaplastic histology was found in hemispheric ependymomas and was associated with older age. EOR and tumor grade were prognostic factors for PFS and OS on multivariate analysis. STR or WHO Grade III pathology, or both, predicted worse overall prognosis in patients with hemispheric ependymoma.
To assess whether a new method of quantifying therapy-associated hemodynamic alterations may help to distinguish pseudoprogression from true progression in patients with high-grade glioma.
Patients ...with high-grade glioma received concurrent chemoradiotherapy. Relative cerebral blood volume (rCBV) and blood flow (rCBF) maps were acquired before chemoradiotherapy and at week 3 during treatment on a prospective institutional review board-approved study. Pseudoprogression was defined as imaging changes 1 to 3 months after chemoradiotherapy that mimic tumor progression but stabilized or improved without change in treatment or for which resection revealed radiation effects only. Clinical and conventional magnetic resonance (MR) parameters, including average percent change of rCBV and CBF, were evaluated as potential predictors of pseudoprogression. Parametric response map (PRM), an innovative, voxel-by-voxel method of image analysis, was also performed.
Median radiation dose was 72 Gy (range, 60 to 78 Gy). Of 27 patients, stable disease/partial response was noted in 13 patients and apparent progression was noted in 14 patients. Adjuvant temozolomide was continued in all patients. Pseudoprogression occurred in six patients. Based on PRM analysis, a significantly reduced blood volume (PRM(rCBV)) at week 3 was noted in patients with progressive disease as compared with those with pseudoprogression (P < .01). In contrast, change in average percent rCBV or rCBF, MR tumor volume changes, age, extent of resection, and Radiation Therapy Oncology Group recursive partitioning analysis classification did not distinguish progression from pseudoprogression.
PRM(rCBV) at week 3 during chemoradiotherapy is a potential early imaging biomarker of response that may be helpful in distinguishing pseudoprogression from true progression in patients with high-grade glioma.
Assessment of radiologic response (RR) for brain tumors utilizes the Macdonald criteria 8 to 10 weeks from the start of treatment. Diffusion magnetic resonance imaging (MRI) using a functional ...diffusion map (fDM) may provide an earlier measure to predict patient survival.
Sixty patients with high-grade glioma were enrolled onto a study of intratreatment MRI at 1, 3, and 10 weeks. Receiver operating characteristic curve analysis was used to evaluate imaging parameters as a function of patient survival at 1 year. Both log-rank and Cox proportional hazards models were utilized to assess overall survival.
Greater increases in diffusion in response to therapy over time were observed in those patients alive at 1 year compared with those who died as a result of disease. The volume of tumor with increased diffusion by fDM at 3 weeks was the strongest predictor of patient survival at 1 year, with larger fDM predicting longer median survival (52.6 v 10.9 months; log-rank, P < .003; hazard ratio HR = 2.7; 95% CI, 1.5 to 5.9). Radiologic response at 10 weeks had similar prognostic value (median survival, 31.6 v 10.9 months; log-rank P < .0007; HR = 2.9; 95% CI, 1.7 to 7.2). Radiologic response and fDM differed in 25% of cases. A composite index of response including fDM and RR provided a robust predictor of patient survival and may identify patients in whom RR does not correlate with clinical outcome.
Compared with conventional neuroimaging, fDM provided an earlier assessment of equal predictive value, and the combination of fDM and RR provided a more accurate prediction of patient survival than either metric alone.
To determine whether early assessment of cerebral white matter degradation can predict late delayed cognitive decline after radiotherapy (RT).
Ten patients undergoing conformal fractionated brain RT ...participated in a prospective diffusion tensor magnetic resonance imaging study. Magnetic resonance imaging studies were acquired before RT, at 3 and 6 weeks during RT, and 10, 30, and 78 weeks after starting RT. The diffusivity variables in the parahippocampal cingulum bundle and temporal lobe white matter were computed. A quality-of-life survey and neurocognitive function tests were administered before and after RT at the magnetic resonance imaging follow-up visits.
In both structures, longitudinal diffusivity (λ(‖)) decreased and perpendicular diffusivity (λ(⊥)) increased after RT, with early changes correlating to later changes (p < .05). The radiation dose correlated with an increase in cingulum λ(⊥) at 3 weeks, and patients with >50% of cingula volume receiving >12 Gy had a greater increase in λ(⊥) at 3 and 6 weeks (p < .05). The post-RT changes in verbal recall scores correlated linearly with the late changes in cingulum λ(‖) (30 weeks, p < .02). Using receiver operating characteristic curves, early cingulum λ(‖) changes predicted for post-RT changes in verbal recall scores (3 and 6 weeks, p < .05). The neurocognitive test scores correlated significantly with the quality-of-life survey results.
The correlation between early diffusivity changes in the parahippocampal cingulum and the late decline in verbal recall suggests that diffusion tensor imaging might be useful as a biomarker for predicting late delayed cognitive decline.