Comparison of multiparametric MRI‐based and transrectal ultrasound‐based preplans with intraoperative ultrasound‐based planning for low dose rate interstitial prostate seed implantation

Abstract Purpose Transrectal ultrasound images are routinely acquired for low dose rate (LDR) prostate brachytherapy dosimetric preplanning (pTRUS), although diagnostic multiparametric magnetic resonance imaging (mpMRI) may serve this purpose as well. We compared the predictive abilities of TRUS vs MRI relative to intraoperative TRUS (iTRUS) to assess the role of mpMRI in brachytherapy preplanning. Materials and methods Retrospective analysis was performed on 32 patients who underwent iTRUS‐guided prostate LDR brachytherapy as either mono‐ or combination therapy. 56.3% had pTRUS‐only volume studies and 43.7% had both 3T‐mpMRI and pTRUS preplanning. MRI was used for preplanning and its image fusion with iTRUS was also used for intraoperative guidance of seed placement. Differences in gland volume, seed number, and activity and procedure time were examined, as well as the identification of lesions suspicious for tumor foci. Pearson correlation coefficient and Fisher's Z test were used to estimate associations between continuous measures. Results There was good correlation of planning volumes between iTRUS and either pTRUS or MRI (r = 0.89, r = 0.77), not impacted by the addition of hormonal therapy (P = 0.65, P = 0.33). Both consistently predicted intraoperative seed number (r = 0.87, r = 0.86). MRI/TRUS fusion did not significantly increase surgical or anesthesia time (P = 0.10, P = 0.46). mpMRI revealed suspicious focal lesions in 11 of 14 cases not visible on pTRUS, that when correlated with histopathology, were incorporated into the plan. Conclusions Relative to pTRUS, MRI yielded reliable preplanning measures, supporting the role of MRI‐only LDR treatment planning. mpMRI carries numerous diagnostic, staging and preplanning advantages that facilitate better patient selection and delivery of novel dose escalation and targeted therapy, with no additional surgical or anesthesia time. Prospective studies assessing its impact on treatment planning and delivery can serve to establish mpMRI as the standard of care in LDR prostate brachytherapy planning.

planning and delivery can serve to establish mpMRI as the standard of care in LDR prostate brachytherapy planning. Group study RTOG 98-05. In this multicenter phase II trial, patients with localized prostate adenocarcinoma (PCa) underwent a preplanning TRUS (pTRUS) volume study alone to plan and guide transperineal low dose rate (LDR) permanent seed implant procedures. This study showed good biochemical control rates, favorable toxicity profiles, and overall survival comparable to other brachytherapy, external beam, and surgical series. 1,2 In addition to preplan imaging which is critical for determining the correct seed quantity and activity, intraoperative planning using TRUS (iTRUS) has been shown to provide important, accurate volume and mapping information to enhance seed placement and limit toxicity to nearby organs. 3,4 Application of magnetic resonance imaging (MRI) for identification and diagnosis of PCa dates back over 30 years. 5 While older iterations of prostate MRI technique lacked sensitivity and specificity (particularly for early-stage tumors), 6 MRI performance has rapidly improved as higher resolution imaging has evolved over the past decade and is expected to further improve with sequence optimizations and other 3D resolution applications. In this setting, MRI has emerged as a useful tool for assessing preoperative staging of PCa.
Most recently, multiparametric MRI (mpMRI) sequences (T2weighted, diffusion-weighted, DCE, MR spectroscopic imaging) have been shown to add important functional data to standard cross-sectional findings, motivating the European Society of Urogenital Radiology (ESUR) to publish clinical guidelines for its use in PCa detection and staging. 7,8 Beyond its diagnostic utility, strong evidence is evolving for the role of mpMRI and image fusion as a useful aid in the treatment planning of prostate cancer. 9 mpMRI in addition to TRUS may enable more precise targeting of high-risk intraprostatic regions without unnecessarily increasing dose to surrounding structures, thereby improving local control. 10,11 In this new era of pretreatment MRI volume studies, the standard pTRUS, which causes a fair amount of patient discomfort and requires additional time and staff, may prove to be redundant. Concern, however, has been raised over the consistency between prostate volumes measured on TRUS compared to those from MRI.
While some reports have demonstrated a tendency of preplan MRI (as well as CT) to overestimate prostate volume compared with ultrasound, 12,13 others found MRI to underestimate gland size relative to TRUS. 14,15 In order to further study the ability of MRI-based preplanning to reliably predict the intraoperative TRUS-based parameters for LDR brachytherapy, we prospectively performed a series of dosimetric preplans using both pTRUs and mpMRI during our transition from pTRUS-to mpMRI-based planning. Specifically, we compared both pTRUS-only-planned studies and MRI-planned studies with a final iTRUS plan to determine the frequency and magnitude of dosimetric changes. The impact of volumetric variation was quantified through changes in the total seed activity necessary and, therefore, in the number of seeds required. The burden of MRI/TRUS fusion was assessed through changes in total procedure and anesthesia time.

| METHODS
Institutional Review Board (IRB) approval was obtained for a retrospective review of our prospectively planned patients with localized prostate cancer treated who underwent LDR permanent seed implantation at our institution from September 5, 2012 to September 6, 2013. Thirty-two patients underwent LDR permanent seed implantation during the study period, all of whom received a pTRUS volume study from which the quantity of seeds and total activity were determined. During the transition to MRI-based preplanning, an additional mpMRI was performed on 14 of these patients (43.7%).
All preplan imaging was acquired in the department of radiation oncology, overseen by a single radiation oncologist who specializes in brachytherapy. The same radiation oncologist, assisted by a single certified medical physicist, performed the brachytherapy preplanning, iTRUS planning, and seed implantation. Pretreatment volume studies were acquired 2-3 weeks prior to the brachytherapy procedure. pTRUS was performed using a stan- Patients were placed in the dorsal lithotomy position and pTRUS images were acquired at 5 mm spacing. These cross-sectional images were used to delineate the required contours and generate a pTRUS-based dosimetric plan. For treatment planning with MRI, multiparametric sequences (T1, T2, dynamic contrast-enhanced series (DCE), diffusion-weighted imaging (DWI)) were acquired through the pelvis with patients in the supine position, and the T2 sequence was used to derive the prostate volume and the number of brachytherapy seeds, from which total seed activity was determined. Multiparametric sequences were used to radiographically identify regions highly suspicious for tumor foci based on the Prostate Imaging Reporting and Data System (PIRADS) scoring system. 8 Typically, a combination of the T2-weighted series then verified on DCE and DWI were referenced to identify foci of disease.
Treatment planning utilizing mpMRI was conducted using MIM planning software to facilitate automated target and normal tissue volume transfers from diagnostic MRI planning to the brachytherapy MIM Symphony program in conjunction with TRUS imaging. Automated MRI-US rigid co-registration, with an estimated 1-2 mm-associated registration error, allowed for more accurate intraoperative adjustments from the initial volume study to further refine treatment dosimetry. A manual readjustment/alignment was also at the operator's disposal to improve fusion accuracy. imaging methods for agreement with the iTRUS treatment. The difference of two coefficients was examined by Fisher's Z transformation test. A P-value of < 5% was considered statistically significant.
Both pTRUS-based and MRI-based preplans accurately and consistently predicted the intraoperative prostate volume. The mean differences between iTRUS vs pTRUS and iTRUS vs MRI were 5 ± 4 cc (P = 0.60) and 4 ± 4 cc (P = 0.58), with correlation coefficient r values of 0.89 and 0.77, respectively (Fig. 1). The between sample difference was similarly nonsignificant (P = 0.97). The mean percent differences between iTRUS vs pTRUS and iTRUS vs MRI were 15 ± 12% and 12 ± 11%, respectively. The predictive abilities of pTRUS and MRI were comparable as well in the subset of patients who had received prior HT, with mean differences between iTRUS vs pTRUS and iTRUS vs MRI of 4 ± 3 cc (P = 0.65) and 4 ± 5 cc (P = 0.33), respectively. The intersample difference was also nonsignificant (P = 0.75). The mean percent differences between iTRUS vs pTRUS and iTRUS vs MRI were 12 ± 8% and 12 ± 7%, respectively. When stratifying for HT use, the difference in volume for the preplanning modalities was nonsignificant (P = 0.35).
In 11 (79%) cases, for which an mpMRI was obtained, a region highly suspicious for focal disease was detected and noted on final radiology read as a PIRADS 4 or 5. None of these biopsy-provendominant tumor foci were retroactively seen on the corresponding pTRUS series. In all cases, the lesions were contoured in the mpMRIpreplans as well as in the intraoperative plans and specifically targeted with dose escalation to 200% of prescription (Fig. 2). This was accomplished through intraoperative MRI/TRUS fusion and real-time planning. All of these index lesions were located, at least in part, in the peripheral zone of the gland, and mpMRI allowed confident exclusion of any concern for extraprostatic extension. When reviewed retrospectively, as above, this distinction was not apparent on TRUS imaging, nor was accurate delineation of zonal anatomy.
There was no significant difference observed in each of the modality's ability to accurately predict the number of seeds that would be required for the brachytherapy implant (Table 2) 28 We demonstrated no difference in mean anesthesia time between pTRUS-alone-planned cases and mpMRI-planned cases.
The application of MRI in brachytherapy treatment planning has become more routine in treatment planning of pelvic malignancies, both in the realm of gynecologic 29 and genitourinary tumors. 8,23 Implementation of new treatment techniques is now possible due to MRI treatment planning. 30 which included 2350 cases of prostate focal therapy, revealed excellent short-and medium-term tumor control and incontinence and erectile dysfunction rates. 34 Part of the impetus to investigate MRIbased partial prostate therapy is mounting evidence that despite the potential for small regions of multifocal disease, it is the primary index lesion representing the most aggressive clonal population that carries clinical significance for potential progression and local disease recurrence. 35,36 Of course, these methods of image-guided targeted biopsy to inform targeted treatment are imperfect and require continued investigation and improvement, an area that we continue to actively investigate. It has become our institutional practice to obtain a diagnostic/planning mpMRI prior to all prostate radiation treatment, including brachytherapy, photon and proton external beam radiation, stereotactic radiotherapy and partial prostate treatment on protocol.
As illustrated in this experience (Fig. 2), any PIRADS 4 or 5 lesion identified on mpMRI is specifically contoured during preplanning for the purposes of focal dose escalation to 200% of prescription.
Limitations of our study include its retrospective nature, although in this context, with the exception of MRI-incompatible implanted hardware, there are no patient or tumor characteristics that would lend bias to undergoing pTRUS alone vs MRI-based preplanning. An additional limitation is the relatively modest sample size which limits our statistical power to detect statistical differences in outcome measures. The sample size, however, is comparable to that of other reported experiences in this realm, and is largely influenced by the adoption of MRI-guided brachytherapy planning for all patients in our department. Analysis of our experience as we considered transitioning from TRUS-based to mpMRI-based preplanning informed the evolution of our institutional practice. Finally, some concern has been raised regarding the accuracy of TRUS/MRI fusion in the operating room due to ultrasound probe deformation of the rectum and secondarily the prostate, and that this anatomical difference may explain any variation between TRUS and MRI imaging. The MIM symphony platform accounts for these changes with fairly high fidelity, allowing for placement of a simulated virtual rectal probe at the appropriate anatomic angle, as well as rigid and deformable image registration. Real-time TRUS guidance is applied during the brachytherapy procedure itself as a standard procedure for final assurance of quality and accuracy.

| CONCLUSION
Our analysis provides further support for mpMRI-only-based LDR