MRI for Assessing Prostate Cancer Response

Twenty patients who are diagnosed with prostate cancer and will undergo external beam radiotherapy will be recruited in this study. Three MR scans of each patient under the study will be performed at different time points. Each MR scan will include several advanced MR imaging sequences (including but not limited to MR spectroscopy, diffusion and perfusion imaging, which are used in routine clinical radiology settings) in addition to typical T1 and T2 weighted MR imaging. The MR scans will be conducted at Radiation Oncology department with the position similar to the treatment position.

The first MR scan will happen prior to the radiation treatment during the scheduled patient simulation for baseline information. The T1- and T2-weighted MRI will be used for delineation of prostate gland as a part of routine treatment planning process. Additional functional MR (DWI/ADC. DCE, etc) images will be used as baseline of this study. The second MR scan with the same sequences will happen in the mid course of radiation treatment. And the third MR scan with the same sequences will happen at end of radiation treatment. Because the participated patients will come to radiation oncology department daily for their routine radiation treatment there will be no additional visits required for patients. The first scan will happen during the same day when patients undergo CT and MRI simulation for treatment planning as part of their standard cares. Two extra MRI scans will be performed as part of this research protocol. The second and third scans will happen on one of treatment days.

While it is anticipated that the advanced MRI techniques will eventually play an important role in the early intervention of radiation treatment and patient management of radiotherapy of prostate cancer at JHH, no modification of the radiation treatment and patient management will be based on the imaging information acquired during this feasibility study. All participated patients will undergo the routine radiation treatment as all other prostate cancer patients managed in our department.

The imaging data acquired will be systematically processed and analyzed along with correlating to radiation treatment dose distribution. All MRI data from three different scans need to be registered to each other and then register to planning CT. The anatomical, functional changes of tumor and normal tissues during radiation treatment will be accessed and correlated to radiation dose. After radiotherapy, all enrolled patients will be followed up the same way as normal post treatment management. The short term or long term tumor response and normal tissue side effects will also be collected under routine clinical follow up.

Following MRI based biomarker will be assessed:

ADC map
Dynamic contrast imaging (DCE)
Spectroscopy
Other MRI modalities as they are developed For example, tumor size measurement, MRI parameters of choline peak in spectroscopy, ADC values, and DCE map, tumor and normal tissue doses will be recorded for each scans. A multi-parametric analysis will be performed to seek correlations.

A new software package will be developed to help overlay the functional MRI data on top of 3D radiation dose distribution. The region of interests and the correlations between MRI data and dose will be easier to identified using under the new software. Further detail quantitative analysis can then be focused on those identified regions of interest.

Data Analysis Steps:

After first scan, dynamic contrast enhanced (DCE) MRI data will be processed by iCAD package to get pharmacokinetic parameters e.g. Ktrans and Ve.. A Ktrans vs Ve map will be generated. All MRI data including T1, T2, ADC map, DEC map, spectroscopy will be resampled to same resolution and registered to each other. The treatment target, prostate in this case, along with surrounding normal organs (rectum, urethral and bladder) will be delineated based on anatomical MRI images. Within prostate gland, different part of regions, cancerous or healthy regions based on functional MRI will be defined too.
Second and third sets of MRI scan data will be processed the same way as the Step 1 and registered to baseline scan. Same anatomical structures and same regions within prostate will be defined.

To test Hypothesis one, the MRI signal from the second and third scan will be compared to the baseline scan on each individual MRI sequence data and on combined all data in a multiparametric approach.The signal changes within different organs and different regions of prostate will be analyzed. A statistical significant (p<0.05) signal intensity changes within certain regions of interest can be defined as observation of signal changes.

After kinetic and other MRI data analysis are done independently without the knowledge of the radiation dose, all registered MRI data will be imported into radiation treatment planning system. Some special software will be developed to change the some MRI data format to a recognized format for planning system. Within the treatment planning system, the MRI data will be co-registered to planning CT. The 3D dose distribution can then overlap on MRI data. The dosimetric information like maximum dose, mean dose of each organ or region of interest defined in previous steps will be calculated. The dose parameters will be calculated from stationary planning CT. There is potential dose uncertainty due to motion of prostate and rectum. However, given that external beam radiotherapy of prostate usually lasts 6-8 weeks with many fractionated treatments and image guided patient treatment ability in our clinic, the systematic organ motion is very small. Besides, most of the literature data on dose dependent tumor control and toxicity were derived from planning CT without considering organ motion. Therefore, we will not consider the effect of organ motion and use the dose parameters from stationary planning CT in this protocol.

The correlation between the dosimetric parameters and the signal changes will be analyzed to test the Hypothesis two.
Routine follow up data on tumor response (e.g PSA value) and normal tissue toxicity assessment will be collected. Although with the limited sample size of this pilot study, the direct link between MRI signal changes and treatment outcome may not be obtained. Any information, trend gathered from the correlation can be used to design next phase study.