Biological Imaging for Optimising Clinical Target Volume (CTV) and Gross Tumour Volume (GTV) Contouring in Prostate Cancer to Improve the Possibilities for Intensity Modulated RadioTherapy (IMRT) Dose Escalation

Proposal:

To investigate the possibilities to improve Clinical Target Volume (CTV) and Gross Tumour Volume (GTV) delineation by using the latest biological imaging modalities on 15 patients with prostate cancer. Our ultimate goal is to set new GTV and CTV definitions and redefine the choice of irradiation margins in Intensity Modulated RadioTherapy (IMRT) for prostate cancer. Furthermore, try to translate the functional imaging data into a Tomotherapy IMRT plan.

Rationale:

Treatment results after standard dose external beam irradiation of locally advanced prostate cancer are insufficient. According to RTOG-8531, RTOG-8610 and EORTC series on T1 to T4 tumours, 5-year overall survival ranges 60 - 73% and 5-year disease free survival ranges 15 - 67%.

Local control can be enhanced by adjuvant androgen suppression, the 5-year disease free survival increases significantly to 36 - 89%. Unfortunately, androgen suppression significantly deteriorates quality of life.

Increasing the irradiation dose also improves local control. However, local toxicity, especially rectal and bladder complications, restrict the dose which can be given with conformal external beam irradiation with population based uncertainty margins. Setup inaccuracy and organ movement determine the irradiation margins needed. Modern position verification techniques, e.g. using fiducial markers in the prostate in combination with megavoltage imaging techniques, allow a reduction of the margins and offer the possibility for dose escalation in the prostate. Studies of IMRT have utilised these sophisticated position verification techniques and this approach appears feasible. A further reduction of the margins, and thus a possible further increasing of the irradiation dose, can be expected from improving the delineation of the prostate contour. Imaging can be performed to delineate anatomic structures or biological processes within the intraprostatic malignant lesion.

CT is commonly used for anatomy delineation, based on early studies. Although the image quality has improved gradually, unfortunately, CT still severely overestimates the prostate volume. Furthermore, Teh et al found in 712 prostatectomy patients large mean differences between CT-based estimates of the GTV and PTV and Pathological Prostate Volume (PPV). Rasch et al found an average ratio between CT and MRI volumes of 1.4. Also the MRI technique improved and now produces a much better imaging quality, e.g. by increasing Tesla, by using the present phased array coils and by using thinner slice thickness. The accuracy of detecting extracapsular extension in prostate carcinoma with endorectal and phased-array coil MR imaging reaches 77% for experienced readers. Concluding, MRI is superior to CT in GTV contouring, but MRI alone (using combined T1 SE and T2 TSE) may not be sufficient for visualisation of the GTV. It is expected that biological imaging will improve GTV delineation further. Within the GTV a Biological Target Volume (BTV) will become visible, which may further improve the efficacy of cancer radiotherapy.

In biological imaging the use of the different imaging techniques have yet to be explored. Currently, there are two topics of importance. The first topic is perfusion. In prostate cancer, the degree of vascularisation appears to correlate with aggressive behaviour and risk of metastasis. In prostate cancer approximately two times as many microvessels exist in the malignant tissue compared to normal tissues. Furthermore, in benign tissues the capillaries are restricted for the most part to the periglandular stroma immediately adjacent to the epithelium, whereas the distribution in carcinoma appears to be more random. Differences in perfusion have been shown to correlate with active prostate cancer areas. Dynamic contrast-enhanced MRI is able to show the microvessel density (MVD) in the prostate. Second is the metabolic activity in prostate cancer. The ratios of choline and creatine (normal value 0.22 +/- 0.13 ppm) reveals metabolic active prostate cancer tissue. Increased choline and/or a reduced citrate indicate prostate cancer. The ratio of choline and creatine-to citrate is related to the Gleason score of the tumour. Concluding, MRI and MRS allow combined structural and metabolic evaluation of prostate cancer location, aggressiveness and stage. The same could be performed using combined CT and PET. Sutinen et al found a clear evidence for detecting areas with [11C] choline in prostate cancer using PET. To our knowledge no data exist comparing [11C] choline PET and MRS in prostate cancer. The technology for biological imaging remains in evolution, and continued advances in accuracy and can be expected.

Therefore, given the current inadequacies in the state of art of defining GTV and CTV in prostate cancer, we propose a study on prostate visualisation using the latest anatomical and biological (metabolic) imaging modalities. Combining the information of different imaging techniques by introducing a BTV and through image fusion is likely to improve CTV and GTV delineation. This will allow us to redefine the choice of margins. Finally, this will result in an improved IMRT plan, where dose painting and dose escalation of the GTV is the ultimate goal. This complies with the current opinion on radiation development. Progress in functional imaging is likely to provide the tools required for individualised risk-adapted radiotherapy.

By introducing a BTV within the GTV a non-uniform CTV delineation can be used and thus margins can be minimised. This allows further escalation of the dose. Setup inaccuracy and organ movement further determine the irradiation margins needed. Therefore, controlling day to day position variability together with the delivery of optimised conformal irradiation will be the next goals to set. Helical Tomotherapy is a novel approach to the delivery of radiation for cancer treatment, which will be able to do both. It relies on a 6 MV linear accelerator for treatment purposes and a 3.5 MV-CT scan for imaging purposes. Both are mounted on a ring gantry that rotates around the patient as he advances through the ring. A 64-leaf collimator defines the radiation fan beam. In a theoretical study Helical Tomotherapy plans required minimal operator interaction and resulted in excellent sparing of normal structures in prostate IMRT. Therefore, in this study we also propose to fuse the anatomic an biologic imaging data with a Tomotherapy MV-CT and make a inverse IMRT Tomotherapy plan. This planning exercise will precede a feasibility study on functional imaging and individualised day to day adapted radiotherapy by Tomotherapy.

Hypotheses:

• It will be possible, using biological (MRS) imaging data and anatomical (CT and MRI) imaging data, to define a BTV within the GTV.
• By introducing a BTV within the GTV a non-uniform CTV delineation can be used and margins can be minimised.
• It will be possible to translate the fused anatomic and biologic imaging data into a clinically sufficient Tomotherapy IMRT plan.

Patients and methods:

Patients who are to receive external beam irradiation for prostate cancer (Stage T1-4 N0/x M0), preferably not treated with anti-androgens and without metal hip prosthesis will be approached to participate in this pilot study. The study will be performed on 15 patients who will receive imaging additionally to the irradiation treatment.

Before start of treatment patients will receive a 3T MRI (T1 SE and T2 TSE), an MRS determining the choline creatine levels in the prostate. Furthermore, an MV-CT will be made as a prelude to future position verification studies. All images (CT, MRI and MRS) will be matched. These combined data will be used to determine an optimal Tomotherapy IMRT plan.

Patients will be treated according to the current department protocol. This study is only an imaging / treatment planning study. No changes in treatment will be made based on the obtained imaging data set. The risk for the patients will be negligible. From the MV-CT approximately 1 cGy. The MRI will be performed without contrast (so no dynamic gadolinium enhanced MRI to evaluate perfusion distribution) to limit the patient burden and risk. Regarding the total irradiation dose of 7200 till 8200 cGy to the patient this additional risk is negligible. The current waiting time for the patient to start with radiotherapy is approximately 5 to 6 weeks. In this waiting time the imaging studies will be performed, so participation to the study will not result in any treatment delay for the patient. The MV-CT scan will take approximately 30 minutes to perform and the combined MRI/MRS approximately 1 hour.

The Cross Cancer Institute in Edmonton facilitates all necessary imaging techniques: a 3T MRI, MRS and a Helical Tomotherapy Unit (TomoTherapy Inc., Madison).

It is not necessary to determine specificity and accuracy of the imaging modalities. The imaging here is only meant to assist in determining the metabolic and anatomic tumour areas in the prostate. Missing a tumour part is not a problem yet because the whole prostate is being treated to the minimal required dose. Overdosage on prostate tissue is not a clinical problem.