Greetje Groenendaal

Validation of functional imaging with pathology for tumor delineation in the prostate

INTRODUCTION A study was performed to validate magnetic resonance (MR) based prostate tumor delineations with pathology. MATERIAL AND METHODS Five patients with biopsy proven prostate cancer underwent a T2 weighted (T2w), diffusion weighted MRI (DW-MRI) and dynamic contrast-enhanced MRI (DCE-MRI) scan before prostatectomy. Suspicious regions were delineated based on all available MR information. After prostatectomy whole-mount hematoxylin-eosin stained (H&E) sections were made. Tumor tissue was delineated on the H&E stained sections and compared with the MR based delineations. The registration accuracy between the MR images and H&E stained sections was estimated. RESULTS A tumor coverage of 44-89% was reached by the MR based tumor delineations. The application of a margin of approximately 5mm to the MR based tumor delineations yielded a tumor coverage of 85-100% in all patients. Errors created during the registration procedure were 2-3mm, which cannot completely explain the limited tumor coverage. CONCLUSIONS An accurate tissue processing and registration method was presented (registration error 2-3mm), which enables the validation of MR based tumor delineations with pathology. Reasonable tumor coverage of about 85% and larger was found when applying a margin of approximately 5 mm to the MR based tumor delineations.

Pathologic Validation of a Model Based on Diffusion-Weighted Imaging and Dynamic Contrast-Enhanced Magnetic Resonance Imaging for Tumor Delineation in the Prostate Peripheral Zone

PURPOSE For focal boost strategies in the prostate, the robustness of magnetic resonance imaging-based tumor delineations needs to be improved. To this end we developed a statistical model that predicts tumor presence on a voxel level (2.5×2.5×2.5 mm3) inside the peripheral zone. Furthermore, we show how this model can be used to derive a valuable input for radiotherapy treatment planning. METHODS AND MATERIALS The model was created on 87 radiotherapy patients. For the validation of the voxelwise performance of the model, an independent group of 12 prostatectomy patients was used. After model validation, the model was stratified to create three different risk levels for tumor presence: gross tumor volume (GTV), high-risk clinical target volume (CTV), and low-risk CTV. RESULTS The model gave an area under the receiver operating characteristic curve of 0.70 for the prediction of tumor presence in the prostatectomy group. When the registration error between magnetic resonance images and pathologic delineation was taken into account, the area under the curve further improved to 0.89. We propose that model outcome values with a high positive predictive value can be used to define the GTV. Model outcome values with a high negative predictive value can be used to define low-risk CTV regions. The intermediate outcome values can be used to define a high-risk CTV. CONCLUSIONS We developed a logistic regression with a high diagnostic performance for voxelwise prediction of tumor presence. The model output can be used to define different risk levels for tumor presence, which in turn could serve as an input for dose planning. In this way the robustness of tumor delineations for focal boost therapy can be greatly improved.

Simultaneous MRI diffusion and perfusion imaging for tumor delineation in prostate cancer patients

BACKGROUND AND PURPOSE A study was performed to investigate if we can quantify if the two imaging modalities diffusion weighted imaging (DWI) and dynamic contrast-enhanced (DCE)-MRI are consistent in what voxels they determine as being suspicious of tumor tissue. MATERIAL AND METHODS Twenty-one patients with biopsy proven prostate cancer underwent a DWI and a DCE-MRI scan. These scans were compared using a receiver operating curve (ROC) analysis, where either one of the two imaging modalities was thresholded and taken as a reference. The resulting area under the curve (AUC) reflects the consistency between target delineations based on the two imaging techniques. This analysis was performed for the complete prostate and the peripheral zone (PZ). RESULTS Consistency between DWI and DCE-MRI parameter maps varied greatly between patients. Values of the AUC up to 0.90 were found. However, on average AUC values were 0.60. The AUC values were related to the patients PSA and clinical stage. CONCLUSIONS Large variation in consistency between the two imaging modalities was found. This did not depend on the precise thresholds used. For making decisions on dose painting in the prostate, the knowledge about the inconsistency must be taken into account.

Phase-based arterial input function measurements in the femoral arteries for quantification of dynamic contrast-enhanced (DCE) MRI and comparison with DCE-CT

Dynamic contrast‐enhanced (DCE) MRI is useful for diagnosis, treatment monitoring and follow‐up of prostate cancer. However, large differences have been reported in the parameter range of the transfer constant Ktrans, making longitudinal studies and comparison of DCE‐MRI findings between studies difficult. Large part of this inconsistency in Ktrans values can be attributed to problems with the accurate measurement of the arterial input function (AIF) from the magnitude signal (AIFMAGN). Phase‐based AIF measurements (AIFPHASE) have been proposed as a more robust alternative to AIFMAGN measurements. This study compares AIFPHASE with AIFs measured with DCE‐CT (AIFCT), and the corresponding Ktrans maps in 12 prostate cancer patients. The shape of AIFPHASE and AIFCT are similar, although differences in the peak height and peak width exist as a result of differences in injection protocol. No significant differences in Ktrans values were found between the DCE‐MRI and DCE‐CT exams, with median Ktrans values of 0.10 and 0.08 min−1 for healthy peripheral zone tissue and 0.44 and 0.36 min−1 for regions suspected of tumor respectively. Therefore, robust quantification of Ktrans values from DCE‐MRI exams in the cancerous prostate is feasible with the use of AIFPHASE. Magn Reson Med, 2011.

Accurate prostate tumour detection with multiparametric magnetic resonance imaging: dependence on histological properties.

Abstract Background. To benefit most of focal treatment of prostate tumours, detection with high precision of all tumour voxels is needed. Although diffusion-weighted imaging (DWI) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) have good diagnostic performance, perfect tumour detection is challenging. In this study, we investigated the variation in prostate tissue characteristics Gleason score (GS), cell density (CD) and microvessel density (MVD) to explain the limitations in tumour voxel detection with a MRI-based logistic regression model. Material and methods. Twelve radical prostatectomy patients underwent a pre-operative 3.0T DWI and DCE-MRI exam. The MRI scans were used to calculate voxel-wise tumour probability with a logistic regression model for the peripheral zone (PZ) of the prostate. Tumour probability maps were correlated and validated with whole-mount histology. Additionally, from the whole-mount histological sections CD, MVD and GS were retrieved for every single voxel. GS, CD and MVD of true- and false-positive voxels and of true- and false-negative voxels were compared using Mann-Whitney U-tests. Results. False-negative tumour voxels had significantly lower CD and MVD (p < 0.0001) and were similar to non-tumour PZ. True-positive detected tumour voxels had high CDs and MVDs (p < 0.0001). In addition, tumour voxels with higher GS showed a trend towards more frequent detection (p = 0.06). Tumour voxels with GS ≥ 3 + 4 showed higher CD and MVD compared to tumour voxels with GS 3 + 3 (p < 0.0001). Conclusion. Tumour voxels with low CD and MVD resemble healthy tissue and are limiting tumour voxel detection using DWI and DCE-MRI. Nevertheless, the most aggressive tumour voxels, containing high CD, MVD and GS, are more likely to be detected and can therefore be treated with high dose using focal therapy or focal boosting.

Dynamic Contrast-enhanced CT for Prostate Cancer: Relationship between Image Noise, Voxel Size, and Repeatability

PURPOSE To evaluate the relationship between image noise, voxel size, and voxel-wise repeatability of a dynamic contrast agent-enhanced (DCE) computed tomographic (CT) examination for prostate cancer. MATERIALS AND METHODS This prospective study was approved by the local research ethics committee, and all patients gave written informed consent. Twenty-nine patients (mean age, 69.1 years; range, 56-80 years) with biopsy-proved prostate cancer underwent two DCE CT examinations within 1 week prior to radiation therapy. Parameter maps of transfer constant (K(trans)), the fraction of blood plasma (v(p)), the fraction of extravascular extracellular space (v(e)), and the flux rate constant between the extravascular extracellular space and plasma (k(ep)) were calculated at 15 different image resolutions, with kernel sizes ranging from 0.002 to 2.57 cm(3). Statistical analysis to quantify the voxel-wise repeatability was performed by using a Bland-Altman analysis on all tracer kinetic model parameter maps of each patient. From this analysis, the within-voxel standard deviation (wSD) was calculated as a function of spatial resolution. RESULTS A kernel size in the range of 0.1-0.3 cm(3) yields reliable information. At 0.15 cm(3), the median wSDs of K(trans), k(ep), v(p), and v(e) are 0.047 min(-1), 0.144 min(-1), 0.011, and 0.104, respectively. With increasing kernel size, these values reach stable levels of approximately 0.02 min(-1), 0.05 min(-1), 0.005, and 0.05, respectively. CONCLUSION There is a high voxel-wise repeatability of the DCE CT imaging technique for prostate cancer for kernel sizes as small as 0.1 cm(3). With the relationship between kernel size, image noise and voxel-wise repeatability, it becomes possible to estimate for alternative DCE CT protocols (eg, those with a reduced radiation dose) at what kernel size a sufficient repeatability can be obtained.

The use of probability maps to deal with the uncertainties in prostate cancer delineation

BACKGROUND AND PURPOSE The use of dynamic contrast-enhanced (DCE) imaging for delineation of prostate tumors requires that decisions are made on a voxel wise basis about the presence of tumor. While the sensitivity and specificity of this technique is high, we propose a probabilistic approach to deal with the intrinsic imaging uncertainty. MATERIAL AND METHODS Twenty-nine patients with biopsy-proven prostate cancer underwent a DCE-CT exam prior to radiotherapy. From a logistic regression on K(trans) values from healthy and diseased appearing prostate regions we obtained a probability function for the presence of tumor. K(trans) parameter maps were converted into probability maps and a stratification was applied at the 5% and 95% probability level, to identify low-, intermediate-, and high-risk areas for the presence of tumor. RESULTS In all patients, regions with high-, intermediate-, and low-risk were identified, with median volume percentages of 7.6%, 40.0%, and 52.1%, respectively. The contiguous areas that resulted from the voxel wise stratification can be interpreted as GTV, high-risk CTV, and CTV. CONCLUSIONS K(trans) parameter maps from a DCE-CT exam can be converted into probability maps for the presence of tumor. In this way, the intrinsic uncertainty that a voxel contains tumor can be incorporated into the treatment planning process.

Why prostate tumour delineation based on apparent diffusion coefficient is challenging: An exploration of the tissue microanatomy

Abstract Background. Focal boosting of prostate tumours to improve outcome, requires accurate tumour delineation. For this, the apparent diffusion coefficient (ADC) derived from diffusion-weighted MR imaging (DWI) seems a useful tool. On voxel level, the relationship between ADC and histological presence of tumour is, however, ambiguous. Therefore, in this study the relationship between ADC and histological variables was investigated on voxel level to understand the strengths and limitations of DWI for prostate tumour delineation. Material and methods. Twelve radical prostatectomy patients underwent a pre-operative 3.0T DWI exam and the ADC was calculated. From whole-mount histological sections cell density and glandular area were retrieved for every voxel. The distribution of all variables was described for tumour, peripheral zone (PZ) and central gland (CG) on regional and voxel level. Correlations between variables and differences between regions were calculated. Results. Large heterogeneity of ADC on voxel level was observed within tumours, between tumours and between patients. This heterogeneity was reflected by the distribution of cell density and glandular area. On regional level, tumour was different from PZ having higher cell density (p = 0.007), less glandular area (p = 0.017) and lower ADCs (p = 0.017). ADC was correlated with glandular area (r = 0.402) and tumour volume (r = −0.608), but not with Gleason score. ADC tended to decrease with increasing cell density (r = −0.327, p = 0.073). On voxel level, correlations between ADC and histological variables varied among patients, for cell density ranging from r = −0.439 to r = 0.261 and for glandular area from r = 0.593 to r = 0.207. Conclusions. The variation in ADC can to a certain extent be explained by the variation in cell density and glandular area. The ADC is highly heterogeneous, which reflects the heterogeneity of malignant and benign prostate tissue. This heterogeneity might however obscure small tumours or parts of tumours. Therefore, DWI has to be used in the context of multiparametric MRI.

The effect of hormonal treatment on conspicuity of prostate cancer: Implications for focal boosting radiotherapy

BACKGROUND AND PURPOSE For focal boosting of prostate tumors, three questions are important regarding the use of hormonal therapy. Does prolonged hormonal treatment affect the conspicuity of tumor tissue on diffusion weighted imaging (DWI) and dynamic contrast-enhanced (DCE-MRI) images? Is tumor delineation possible in patients using hormonal treatment? Can we identify specific imaging thresholds for tumor delineation in patients after prolonged androgen deprivation? MATERIALS AND METHODS Ninety-six patients were included. Using multivariate linear regression analyses, we investigated if DWI and DCE-MRI parameter maps are different in patients receiving hormonal treatment for 0-3 or >3 months. Furthermore, logistic regression was performed to obtain specific imaging thresholds for tumor tissue for the two patient groups. RESULTS We found a significantly higher diffusion and lower perfusion of tumor tissue in the >3 months hormonal treatment group compared to the 0-3 group. This resulted in lower tumor conspicuity. Nevertheless, in 18/21 of the patients in the >3 months treatment group, a suspicious lesion could be defined based on the MR images. Based on logistic regression, different imaging thresholds should be set for tumor detection in the two treatment groups. CONCLUSIONS Prolonged androgen deprivation decreases tumor conspicuity. Different imaging thresholds need to be set to delineate tumor in patients who have had prolonged hormonal treatment.

Expression of hypoxia-inducible factor-1α and -2α in whole-mount prostate histology: Relation with dynamic contrast-enhanced MRI and Gleason score

The aim of this study was to investigate the association between the immunohistochemical expression of hypoxia-inducible factor (HIF)-1α and HIF-2α and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parameters Ktrans and kep in prostate cancer. Therefore, 15 patients with biopsy-confirmed prostate cancer underwent a pre-operative 3T DCE-MRI scan. Immunohistochemical analysis of HIF-1α and HIF-2α, and of CD31 for microvessel density (MVD) was performed. Tumor areas were delineated on whole-mount histopathological sections. Nuclear HIF expression was correlated with the quantitative DCE-MRI parameters Ktrans and kep, MVD and Gleason score. HIF expression was highly heterogeneous within tumors and between patients. Pronounced expression of HIF-2α was present, while HIF-1α expression was more limited. Larger tumors showed higher HIF-2α expression (p=0.041). A correlation between HIF-2α and Ktrans p5th was found (r=0.30, p=0.02), but no differences in Ktrans, kep and MVD were observed for different levels of HIF expression. HIF expression was not associated with Gleason score. In conclusion, in this whole-mount prostate cancer study, larger prostate tumors showed frequently high HIF-2α expression, suggesting that larger tumors are clinically most relevant. However, HIF-1α and HIF-2α were not correlated with DCE-MRI parameters. Given the pronounced expression of HIF-2α and independence of Gleason score, HIF expression may function as a biomarker to guide boost dose prescription.