Feasibility of neurovascular sparing MR-guided adaptive radiotherapy for prostate cancer

Abstract

Purpose or Objective Erectile dysfunction is a common adverse effect of external beam radiation therapy for localized prostate cancer (PCa), probably due to damage to surrounding neural and vascular tissue. MR-guided on-line adaptive radiotherapy (MRgRT) enables high-resolution MR imaging during dose delivery and facilitates correction for both inter- and intra-fraction movement and tissue deformations, paving the way for a neurovascular sparing approach to reduce erectile dysfunction after radiotherapy for PCa. Materials and Methods Monaco 5.40 (Elekta AB) was used to generate 5x7.25 Gy neurovascular sparing MR-Linac treatment plans for an unselected consecutive series of 10 localized PCa patients, previously treated with conventional 5x7.25 Gy MRgRT. Seven-field intensity modulated radiation therapy technique was used. The gross tumor volume (GTV) included the MR visible tumor with a 4mm isotropic margin excluding OAR and the planning target volume (PTV) included the GTV and prostate body with a 5mm isotropic margin. In addition to the organs at risk (OAR) such as rectum and bladder, dose constraints for the neurovascular bundles (NVBs), the internal pudendal arteries (IPAs), the corpora cavernosa (CCs), and the penile bulb (PB) were established and dose prescriptions for GTV and PTV were adapted (figure 1). For the treatment planning the primary goal was to achieve clinically acceptable dose coverage for both GTV and PTV, secondary sparing of OAR, and tertiary sparing of neurovascular structures. When constraints of the NVB could not be met, a dose as low as reasonably achievable was pursued. Dose to regions of interest were compared between the neurovascular sparing plans and conventional plans. Results Constraints for the IPAs, CCs, and PB were met in all 10 cases. Constraints for the NVBs were met in 5 cases bilaterally, in 3 cases unilaterally, and were not met in 2 cases (figure 2). In the cases where the NVB constraint was not met, the mean dose was still significantly lower compared to the conventional plans (mean Dmean 27.75 Gy vs. 34.90 Gy). Unfavorable cases for neurovascular sparing were those with a GTV in the dorsolateral position, a wider spread of the NVBs around the prostate, and IPAs running in closer proximity along the prostate. Conclusion Neurovascular sparing MRgRT for localized PCa is feasible. The extent of NVB sparing largely depends on the patient’s anatomy and GTV location. We will initiate a phase II clinical trial to investigate the effect of neurovascular sparing MRgRT on erectile function in intermediate risk PCa patients with a satisfactory erectile function at baseline.