4791

Ex vivo MRI evaluation of prostate cancer: localization and margin status prediction of prostate cancer in fresh radical prostatectomy specimens
Jan Heidkamp1, Martijn Hoogenboom1, Iringo Kovacs2, Andor Veltien1, Arie Maat2, Michiel Sedelaar3, Christina Hulsbergen-van de Kaa2, and Jurgen Fütterer1

1Radiology and Nuclear Medicine, Radboud university medical center, Nijmegen, Netherlands, 2Pathology, Radboud university medical center, Nijmegen, Netherlands, 3Urology, Radboud university medical center, Nijmegen, Netherlands

Synopsis

This study investigated the ability of high field ex vivo MRI to localize prostate cancer (PCa) and to predict the margin status in fresh radical prostatectomy (RP) specimens using histology as gold standard. In twelve specimens, ex vivo MRI localized 17 (47%) of 36 PCa lesions confirmed by histological examination. Ex vivo MRI identified none of the 4 histological positive surgical margins (sensitivity 0%) and 9 of the 13 negative margins (specificity 69%). Our results indicate accurate localization of PCa in fresh RP specimens by ex vivo MRI, yet the technique did not perform well in predicting the margin status.

Introduction

Radical prostatectomy (RP) is the recommended curative treatment in patients with low- and intermediate risk localized prostate cancer (PCa) and a life expectancy of more than 10 years.1 At histological examination, tumor that extends to the inked surface of the prostatectomy specimen is defined as a positive surgical margin (PSM) and is considered an adverse patient outcome.2–4 Currently, indisputable and topical information on the margin status is unavailable during or directly after surgery. Macroscopic evaluation by palpation is unreliable and per-operative frozen section analysis (FSA) is known to suffer from sample errors.5 Histological examination requires at least several days for a definitive result. Until then, neither the surgeon nor the patient can be fully certain of the treatment success. Acquiring ex vivo MR images of the RP specimen might be a way to provide per-operative information on the margin status which could be exploited to intraoperatively direct additional resection. This could potentially decrease the number of PSM found at histological examination and the subsequent need for radiotherapy or repeated surgery. In the present study, the ability of high field ex vivo MRI to localize PCa and to predict the margin status in fresh RP specimens is investigated.

Methods

The Institutional Review Board approved this study and written informed consent was obtained. Patients with biopsy-proved PCa and a multi-parametric MRI examination prior to undergoing RP were prospectively included. Immediately after surgical resection, the entire intact prostate specimen was positioned in a custom made container manufactured from MR compatible acrylic glass that provided a reference between histological slicing and the ex vivo MRI. The container with specimen was positioned in a 7 Tesla preclinical MR scanner interfaced to a Siemens console. Turbo Spin Echo (TSE, TE/TR = 56/4,800, voxel size = 0.13x0.13x2 mm, matrix = 512x512) and half-Fourier acquisition single shot turbo spin echo (HASTE, TE/TR = 44/2,000, voxel size = 0.51x0.51x2 mm, matrix = 128x96) with b-values of 0-100-500-1000-1200 s/mm2 sequences were applied to acquire T2 and diffusion weighted images respectively. Furthermore, apparent diffusion coefficient (ADC) maps were calculated. After the ex vivo MRI regular histological work-up followed. A pathologist and a radiologist separately annotated areas of PCa and the surgical margin status in the histology slides and ex vivo MRI respectively. Any contact of a lesion visible on ex vivo MRI was defined as a PSM. The radiologist determined the presence of PSM with a five-point Likert scale.

Results

In 12 RP specimens, histopathology revealed 36 PCa lesions of which 17 (47%) were correlated with a lesion annotated on the ex vivo MRI. The ex vivo MRI localized 12 of 25 peripheral zone lesions (48%), 5 of 11 transition zone lesions (45%), 10 of 24 Gleason score ≤ 6 lesions (42%), 7 of 12 Gleason score > 6 lesions (58%), 6 of 18 ≤ 0.5 cc lesions (58%), and 11 of 18 > 0.5 cc lesions (61%). 7 of 11 (63%) Gleason score > 6 and > 0.5 cc lesion were localized. The ex vivo MRI identified none of the 4 histological positive surgical margins (sensitivity 0%) and 9 of the 13 negative margins (specificity 69%). The four false positive were assigned an average Likert score of 3.8 and 3.3, the four false negatives 2.0 and 1.8, and the true negatives 1.8 and 1.9 on T2W and DW imaging respectively.

Discussion

To our knowledge, ex vivo MRI has never been applied to localize PCa and predict margin status in fresh RP specimens. This study demonstrated an overall localization rate for PCa of 47% which is comparable with in vivo findings of Tan et al.6 The localization rate might improve with a more accurate correlation between histology and ex vivo MRI. This could be achieved by improving the design of the container limiting the variation of angulation of histological slicing. The fact that a prostatectomy is sometimes performed maintaining a narrow margin (≈ 1 mm) only might have resulted in the poor sensitivity of ex vivo MRI for PSM, despite its high resolution (0.13x0.13x2 mm). The Likert scores for the false positive surgical margin represent uncertainty of the reader.

Conclusion

Ex vivo MRI enabled accurate localization of PCa in fresh RP specimens, yet the technique was unable to provide indisputable information on the margin status. The ability of ex vivo MRI to localize PCa might complement intraoperative frozen section analysis that has a stronger performance to predict surgical margin status

Acknowledgements

Ex vivo MRI was performed using equipment of the Preclinical Imaging Centre (PRIME) within the Radboud university medical center, Nijmegen, the Netherlands.

References

1. Heidenreich A, Bastian PJ, Bellmunt J, et al. EAU guidelines on prostate cancer. Part 1: screening, diagnosis, and local treatment with curative intent - Update 2013. Eur Urol. 2014;65(1):124–137.

2. Epstein JI, Amin M, Boccon-Gibod L, et al. Prognostic factors and reporting of prostate carcinoma in radical prostatectomy and pelvic lymphadenectomy specimens. Scand J Urol Nephrol Suppl. 2005;39(216):34–63.

3. Yossepowitch O, Bjartell A, Eastham J a., et al. Positive surgical margins in radical prostatectomy: outlining the problem and its long-term consequences. Eur Urol. 2009;55(1):87–99.

4. Yossepowitch O, Briganti A, Eastham J a., et al. Positive surgical margins after radical prostatectomy: a systematic review and contemporary update. Eur Urol. 2014;65(2):303–313.

5. Ramírez-Backhaus M, Rabenalt R, Jain S, et al. Value of frozen section biopsies during radical prostatectomy: significance of the histological results. World J Urol. 2009;27(2):227–234.

6. Tan N, Margolis DJ, Lu DY, et al. Characteristics of detected and missed prostate cancer foci on 3 T multiparametric MRI using an endorectal coil correlated with whole-mount thin-section histopathology. Am J Roentgenol. 2015;205(1):87–92.

Figures

Flowchart depicting the employed method for specimen handling. *This hour included transportation from the operation theatre to the pathology laboratory and subsequent transportation to the MRI facility.

Schematic representation of the RP specimen position within the PMMA container. B, base; A, apex; L, left; R, right. (a), side view of the setup, with the arrows representing the saline filled rods. The wooden pins by which the specimen was pinned down are represented by the slightly oblique rods. The solid grey layer represents the moulded layer of paraffin. (b), top view of the setup, with the triangles depicting the saline filled rods and the circles the wooden pins.

For a 67-year-old, (a), the ex vivo T2W images, showing a hypointense lesion in the left TZ suggestive of PCa (T); (b), approximately corresponding histology slide. PCa is annotated by the solid line; (c), ADC map demonstrating diffusion restriction at the location of PCa; (d), annotation by the radiologist. The figure demonstrates a corresponding position of PCa between ex vivo MRI and histology. The band of hyperintense non-pathological PZ tissue separating the PCa lesion and the edge of the specimen visible on the T2W images (white arrowheads) was indicative for a NSM. Histology evaluated this as NSM (black arrowheads).

For a 62-year-old patient, (a), the ex vivo T2W images, showing an area of PCa (T) situated ventrally left in the prostate; (b), approximately corresponding histology slide showing two cancer lesions (dotted lines); (c), ADC map showing a hypointense area of diffusion restriction suggestive for PCa; (d) annotation by the radiologist. On ex vivo MRI this lesion was incorrectly considered as having a NSM on indication of the small strip of tissue separating the lesion and the edge of the specimen visible on the T2W images (white arrowheads). However, histological examination confirmed a PSM (black arrow).

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
4791