Patient Population
The DCVax-L trial at the University of Pennsylvania, “A Phase-3 Clinical Trial Evaluating DC Vax®L, Autologous Dendritic Cells Pulsed with Tumor Lysate Antigen for the Treatment of Glioblastoma Multiforme”, per protocol, NCT00045968, was approved by the Abramson Cancer Center –Scientific Review Committee (UPCC # 34313) and the Institutional Review Board (#81817). The study was designed as a randomized, placebo-controlled, double-blinded, multi-center, multi-national phase-3 clinical trial. The trial randomized patients with newly diagnosed-glioblastoma with no evidence of progression at baseline into the treatment cohort to receive DCVax-L as adjuvant therapy to SOC treatment and into the placebo cohort to receive SOC treatment supplemented with autologous mononuclear cells (MNC; placebo). Patients in either cohort who progressed during the study were offered a crossover (open-label) option with no unblinding to receive DCVax-L.
The inclusion criteria for enrollment in the present study were that all patients i) had histologically confirmed diagnosis of glioblastoma [19]; (ii) had been treated with SOC, i.e., surgery and CCRT, iii) presented with a new enhancing lesion in the radiation field after completion of CCRT, (iv) underwent successful administration of DCVax-L, and (v) had availability of standard and advanced (DTI and DSC) neuroimaging data at follow-up time points.
Based upon the inclusion criteria, a cohort of 17 patients (mean age = 58.1 ± 6.7 years, 13 males/ 4 females) with glioblastoma treated with SOC treatment plus DCVax-L were included in this study. In patients in whom tumor specimen was available from repeat surgery/biopsy, malignant features on histopathology were used to identify TP + mixed response (> 25% viable cells with malignant features; n = 9) or PsP (< 25% malignant features; n = 2) [20]. In the case of non-availability of tissue specimens, ≥ 2 consecutive follow-ups MRI scans using updated mRANO criteria [21] were used to determine the final diagnosis of TP + mixed response (n = 2) or PsP (n = 4). The multiparametric MRI-based prediction model [18] consisted of a combination of DTI-derived fractional anisotropy (FA), coefficient of linear anisotropy (CL), and dynamic susceptibility contrast (DSC)-perfusion MRI-derived maximum relative cerebral blood volume (rCBVmax) parameters from contrast-enhancing lesions in differentiating TP + mixed response from PsP. This model was developed based on the histological analyses of tumor specimens with an accuracy of over 90% in our previous study [18].
Therefore, we used a combination of these three parameters to compute the predictive probabilities (PP) of true progression using the following regression equation:
$$f(FA,CL, rCBVmax)=1÷1+exp(-(\beta 0+\beta 1FA+\beta 2CL+\beta 3rCBVmax\left)\right)$$
,
where β0 = -16.17, β1 = 194.01, β2 = -285.65, and β3 = 1.21.
The most recent MRI (with availability of DTI and DSC sequences) preceding re-resection/biopsy, when progressive disease was suspected, was used to compute PP values in the present cohort. The enhancing lesions were considered as ‘TP + mixed response’ if the PP was ≥ 50% and ‘PsP’ if PP was < 50% [13–15].
The overall survival (OS) was considered the primary clinical endpoint. The survival time of all the patients was recorded from the date of surgery to death. In the present study, a cohort of 48 glioblastoma patients (treated with surgery followed by CCRT) from our previous study [22] was included as an external control group to compare survival outcomes between patients treated with SOC and patients treated with SOC plus DCVax-L. There were no significant differences in age, gender, and KPS between these two groups of patients [23]. All these patients were treated at the same institution undergoing surgery and adjuvant treatment by the same clinical team.
MRI Data Acquisition
All patients underwent MRI on a 3T Tim Trio whole-body MR scanner (Siemens, Erlangen, Germany) using a 12-channel phased-array head coil. The anatomical imaging protocol included axial 3D-T1-weighted magnetization-prepared rapid acquisition of gradient echo (MPRAGE) imaging and an axial T2-FLAIR imaging using standard parameters. The postcontrast T1-weighted images were acquired with the same parameters as the precontrast acquisition after administration of standard dose (0.14 mmol/Kg) of gadobenate dimeglumine (MultiHance, Bracco Imaging, Milano, Italy) intravenous contrast agent using a power injector (Medrad, Idianola, PA). Additionally, DTI and DSC-PWI sequences were acquired using the parameters described previously [18, 24, 25].
Image Processing And Data Analysis
Motion and eddy current correction modules were applied to raw DTI data using an in-house developed software (IDL; ITT Visual Information Solutions, Boulder, Colorado). Pixel-wise mean diffusivity (MD), FA, CL, planar anisotropy (CP), and spherical anisotropy (CS) maps were computed by using the methods reported previously [24, 25]. Leakage-corrected CBV maps were generated by performing gamma-variate curve fitting from DSC-perfusion MRI data using NordicICE software (NordicNeuroLab, Bergen, Norway).
The DTI-derived maps, CBV maps, and T2-FLAIR images were resliced and co-registered to contrast-enhanced T1-weighted images. A semiautomatic approach was used to segment the contrast-enhancing regions of each lesion by using a signal intensity-based thresholding method [24, 25]. The median values of DTI metrics (MD, FA, CL, CP, and CS) from the enhancing regions were measured. In addition, the lower 10th percentile MD values were measured from the enhancing region and were reported as MDmin. The CBV values from the enhancing regions were normalized by corresponding values from contralateral normal white matter to obtain relative CBV (rCBV). The top 90th percentile rCBV values were also measured from the enhancing region and were reported as rCBVmax.
Radiographic Response Assessment Using mRANO Criteria
Patients in whom repeat surgery, or biopsy was not performed, mRANO criteria [21] were used to determine the final diagnosis of TP + mixed response or PsP by a board-certified neuroradiologist (SM).
Histological/immunohistochemical Analysis
Tumor specimens were originally cut, mounted, and stained with hematoxylin-eosin by standard methods. The status of isocitrate dehydrogenase (IDH), O6-methylguanine-DNA-methyltransferase (MGMT), and EGFRvIII was determined using standard procedures [26] by a board-certified neuropathologist (MPN) who was blinded to the neuroimaging results.
Statistical Analysis
Kolmogorov-Smirnov tests were used to determine the nature of data distribution. As the data showed departure from Gaussian distribution, non-parametric Mann-Whitney U tests were performed to assess differences in the median values of MD, FA, CL, CP, CS, rCBV, MDmin, and rCBVmax between TP + mixed response and PsP groups as determined by histopathology/mRANO criteria and by PP values. A probabilistic (p) value of less than 0.05 was considered significant. We sought to ascertain the number of cases correctly classified as PsP or TP + mixed response using multiparametric MRI-based prediction model using histopathology/mRANO criteria as ground truth in the final diagnosis of PsP and TP + mixed response. Pearson test was used to determine the concordance correlation coefficient between PP values and histopathology/mRANO criteria. Kaplan-Meier survival curve plots and log-rank tests were used to compare the OS using stratification factors: i) TP + mixed response and PsP treatment outcomes obtained from multiparametric MRI-based prediction model, and ii) histopathology/mRANO criteria; iii) MGMT profile (methylated and unmethylated status) ;iv) Ki-67 proliferation index in the recurrent specimens (above and below the median Ki-67); and v) imaging parameters (above and below the median values of FA, CL, and rCBVmax). A comparison of OS outcomes was performed between two matched groups of patients, one treated with SOC alone, the other DCVax-L plus SOC. Additionally, Pearson exact χ2 tests were performed to estimate the frequency of MGMT promoter methylation in TP + mixed response and PsP cases. All statistical analyses were performed using a statistical package, SPSS for Windows (v. 18.0; Chicago, IL).