UROLOGICAL ONCOLOGY Multiparametric MRI for the Diagnosis of Tumor Type in Patients Suspicious of Inner Gland Prostate Cancer Zahra Ghane1,2, Fariborz Faeghi3*, Mahyar Ghafoori4, Abolfazl Payandeh5 Purpose: The current study aimed to evaluate multiparametric MRI for the diagnosis of type of tumor (benign or malignant) in patients suspicious of inner gland prostate cancer. Materials and Methods: This cross-sectional study was conducted on 44 consecutive patients with a clinical impression of prostate cancer who were referred to the MRI department of Payambaran Hospital, Tehran, Iran for confirmative diagnostic evaluation. Cases suspected of tumor relapse and those who previously underwent treatment for prostate cancer were excluded. Multiparametric MRI was performed for every patient by using a 1.5 Tesla device with an integrated endorectal and pelvic-phased array coil. All patients subsequently underwent MRI- transrectal ultrasound fusion biopsy. The diagnostic value of each sequence was then investigated individually and in combination with other techniques by comparing the results with histological findings from MRI–TRUS fusion biopsy. Results: Among the techniques, T2-weighted imaging (T2W) had the highest sensitivity and specificity while dynamic contrast enhanced (DCE) technique had the least. Diffusion-weighted imaging (DWI) and magnetic res- onance spectroscopy (MRS) had a similar sensitivity and specificity and did not significantly differ from T2W. Adding functional techniques to T2W did not improve diagnostic indices compared to T2W alone. Quantitative evaluation of apparent diffusion coefficient (ADC), DWI, and MRS showed that all techniques were able to dif- ferentiate between benign and malignant tumors. However, the quantitative combination of these sequences de- creased diagnostic performance. Conclusion: T2W is the best technique for the diagnosis of type of tumor in terms of benignancy or malignancy in patients suspicious of inner gland prostate cancer. Adding functional imaging measurements to T2W does not improve its diagnostic value. Keywords: multiparametric MRI; prostate cancer; zone; T2 weighted imaging. INTRODUCTION Cancer has become a major public health problem and accounts for the third leading cause of death in Iran. Specifically, prostate cancer has turned into an important issue in the world especially in developing countries. It is the second most prevalent cancer in the world and the sixth most prevalent in Iran. The most common histology observed in prostate cancer is ade- nocarcinoma which is also associated with a shorter life span.(1) Regarding zonal origin, 65% of prostate tumors originate from the peripheral zone while about 30% of them develop from the transition zone. The presence of transition zone tumors plays a significant role in the progression and mortality of the disease. Thus, the early diagnosis is essential.(2) Prostate cancer is initially diagnosed by measuring 1Department of Radiology Technology, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran. 2Iran Social security Organization. 3Department of Radiology Technology, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran. 4Radiology Technology Department, Full Professor, School of Allied Medical Sciences, Iran University of Medical Sciences, Tehran, Iran. 5Department of Biostatistics and Epidemiology, School of Health, Zahedan University of Medical Sciences, Zahedan, Iran. *Correspondence: Tehran - Shemiran - Tajrish - Ghods Square - Darband Street - School of Allied Medical Sciences. Postal code: 1971653313, Telefax: +98. 2122722150, Tel: +98.2122718506. Email: f_faeghi@sbmu.ac.ir. Received December 2018 & Accepted October 2019 prostate-specific antigen (PSA) level and performing digital rectal exam (DRE). Definite diagnosis is made through transrectal ultrasound-guided (TRUS) biopsy. However, these diagnostic techniques have some draw- backs. Low sensitivity and low positive predictive val- ue (PPV) of DRE, low specificity of PSA measurement, and inefficacy of systemic biopsy in diagnosing cancers of the anterior part of the prostate are some of the lim- itations related to these methods. Hence, identifying a non-invasive and more precise method for early diag- nosis of prostate cancer is crucial.(3) In the mid-1980s, for the first time T1-weighted (T1W) and T2-weighted (T2W) imaging techniques of MRI were used for prostate imaging. Gradually, by adding functional imaging (DWI, DCE, and MRS) to the ana- tomic sequence (T2W), it became possible to examine Urology Journal/Vol 16 No. 6/ November-December2019/ pp. 552-557. [DOI: 10.22037/uj.v0i0.4998] Vol 16 No 06 November-December2019 553 the physiological properties of tissues. Among these functional parameters, dynamic contrast enhanced (DCE) is efficient for assessing microvascular proper- ties, diffusion weighed imaging (DWI) is sensitive to the restriction of water molecule diffusion movement and magnetic resonance spectroscopy (MRS) is val- uable for evaluating biochemical changes within the prostate tissue. However, none of these functional tech- niques are sufficient for the diagnosis of prostate cancer individually. Moreover, heterogeneous appearance and overlap enhancement of BPH nodules originating from the transition zone complicates the detection of tumors originating from this zone.(3,4) The current study was conducted to evaluate the effica- cy of multiparametric MRI (mpMRI) for the diagnosis of type of tumor in the inner prostate gland (transition, central, and fibromuscular zone). Also, we aimed to compare the results obtained from MRI with the results of MRI-TRUS fusion biopsy as it is considered the golden standard of diagnosis. MATERIALS AND METHODS Study Design, Sample and Population This cross-sectional study was performed during 2017 on patients clinically suspicious of prostate cancer who were referred to the MRI department of Medical Im- aging Center of Payambaran Hospital, Tehran, Iran for further diagnostic evaluation. Sample size was comput- ed by using PASS software (version 11.0.4). A sample size of n=44 was required to achieve 80% power for the two-sided binomial test to detect a change in sensi- tivity from 0.5 to 0.8. The probability of type one error (（) was considered to be 0.05. Also, based on previous studies, the prevalence of prostate cancer was estimated to be 0.50.(5) The present research was approved by the Ethics Committee of Shahid Beheshti University of Medical Sciences, Tehran, Iran (ethics code: IR.SBMU. RETECH.REC.1396.828). Inclusion and exclusion criteria This study included men ≥ 50 years old with a clinical suspicion of prostate cancer who were referred by an urologist for MRI imaging and prostate biopsy. The pri- mary diagnosis was based on an increase in the serum level of PSA (PSA> 3ng/mL) or an abnormal DRE. Ex- clusion criteria included tumor relapse, having already undergone treatment, and having contraindications for receiving endorectal coil such as presence of severe hemorrhoid or severe inflammatory bowel disease, sen- sitivity to latex, or history of rectal resection. Subjects who were contraindicated for MRI imaging (i.e. pres- ence of ferromagnetic implants and cardiac pacemak- ers) or gadolinium contrast agent injection (i.e. active asthma, allergy to gadolinium, severe allergy, and GFR < 30 ml/min) were also excluded from the study. In ad- dition, cases whose obtained images were not satisfac- tory (e.g. multiple artifacts due to total hip replacement or patient movements) were also not included. Procedures In this study, an MRI scanner with a field strength of 1.5 Tesla (Magnetom Avanto, Siemens) along with combined endorectal and pelvic phased-array coils was used. Multiparametric sequences including T2W, DWI, DCE, and MRS were performed for all patients. The detail of each protocol is shown in Table 1. For DCE MRI, 0.1 mmol/kg gadolinium contrast agent was ad- MpMRI for the detection of inner gland prostate cancer-Ghane et al. Pulse sequences T1-W (TRA) T2-W(TRA-SAG-COR) DWI (TRA) DCE (TRA) MRS (3D-CSI) Time Repetition (ms) 600 8000 4400 4.96 650 Time Echo (ms) 12 109 82 1.69 120 (mm)slice thickness 3 3 3 3 - Matrix size 256×192 320×320 102×50 192×138 - number of section 24 24 30 28 10 Field Of View(mm) 175×175 175 ×175 175×85 250×250 190×190 Flip Angle 150 150 - 12 90-180-180 Average (NEX) 1 1 6 1 8 (mm) voxel size 0.9×0.7×3 0.5×0.5×3 1.7×1.7×3 1.9×1.4×3 10×10×10 Temporal resolution - - - 11s - b-value(s/mm2) - - 50-400-800-1200 - - Table 1. Details and parameters of the sequences Figure 1. Classification of DCE I: Slow entrance of the contrast agent and contrast is kept being enhanced and having the same signal with background tissue II: Rapid entrance and relatively rapid exit of the contrast agent from tissue III: Very rapid entrance and exit of the contrast agent from tissue ministered with an injection rate of 2-3 ml/s followed by 20 mL normal saline flush. The temporal resolution was equal to 11 seconds. Evaluations Initially, an experienced radiologist interpreted the MRI images and correlated the lesions observed in T2W with functional sequences. Diagnostic characteristics for the detection of tumoral lesions included lesion morpholo- gy and homogeneous low signal intensity in T2W, re- stricted diffusion in DWI, early enhancement and wash out of the contrast agent in dynamic imaging, and in- crease in the choline + creatinine to citrate ratio (Cho + Cr/Ci) in MR spectroscopy. According to the PI-RADS v2 scoring system, findings of T2W and DWI were as- sessed on a 5-point category scale with 5 being most likely to represent clinically significant prostate cancer. MRS was also assessed with a score from 1 to 5. DCE was evaluated based on the shape of the curves (Fig- ure1 and 2). MRI- TRUS fusion biopsy was considered as the gold standard of diagnosis. Considering the correlation between the imaging scores and the results obtained from fusion biopsy, the scores of 1 and 2 were considered as negative, score 4 and 5 as positive, and the score of 3 was considered negative for T2W and DWI and positive for MRS. As for the DCE technique, asymmetry and focal early enhancement were assumed to be positive along with the shape of plateaus and washout. In the second stage of evaluation, quantitative and semi-quantitative values were obtained using the syngo MRI software (Siemens Medical Solutions), spectros- copy software, and mean curve. Quantitative factor of diffusion coefficient, the ratio of metabolites in MRS and TIC pattern diagrams were considered as quantita- tive and semi-quantitative values. Advanced ultrasound devices that were equipped with special software and hardware to accurately match MRI images with ultrasound images were used for tissue sampling. With the help of the sensors connected to the ultrasound probe and the patient's body, the probe's po- sition relative to the prostate was detected at any time and by moving the probe in different directions within the rectum, an ultrasound image was provided on the monitor as well as an equivalent MRI image. By mark- ing the suspicious mass on the MRI image, the same area was automatically marked on the ultrasound im- age, and the corresponding software specified the nee- dle pathway to obtain a tissue sample from the mass. Then, samples were sent for pathological evaluation. Statistical Analysis Analysis was performed using SPSS software version 20 (IBM, Chicago, Illinois, USA) and MedCalc version 12.1.4 (MedCalc Software bvba, Mariakerke, Belgium). Diagnostic indices including the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, ROC curve, and area under the curve (AUC) of each sequence was calculated separate- ly. McNemar’s test was employed for the comparison of diagnostic values and logistic regression model was used for the evaluation of combined MRI sequences. P-value < .05 was considered as statistically significant. RESULTS A total of 44 male patients suspicious of prostate cancer with a mean age (SD) of 65.1(5.9) years old (range: 53 – 80) and a serum PSA level > 3 ng/mL were included in the study. Based on histological examination, 22 pa- tients (50%) were diagnosed with malignancy with a Gleason Score ≥ 6 and the rest had benign tumors. Regarding the ability of each technique to detect the type of tumor separately, results obtained from imag- ing with T2W, DCE, DWI and MRS (qualitative) se- quences were not compatible with histologic findings in 5,19,9, and 9 cases, respectively. Table 2 reports the diagnostic indices of these sequences when performed in isolation. There was a significant difference between the specificity of DCE and T2W techniques only (P = .004). As shown in Figure 3, T2W had a greater AUC compared to the other techniques. The estimated AUC was equal to 0.89 (P < .001), 0.57 (P = .44), 0.80 (P = .001), and 0.80 (P = .001) for T2W, DCE, DWI, and MRS sequences, respectively. The results for double, triple, and quadruple combina- tions of the aforementioned techniques (T2W+DWI, T2W+MRS, T2W+DCE, T2W+DWI+DCE, T2W+D- WI+MRS, T2W+DWI+DCE +MRS) were found to be similar to those of T2W alone (P > 0.05). In oth- er words, by adding functional sequences to T2W, no changes were observed in terms of diagnostic indices. The techniques of DWI, ADC, and MRS were also investigated quantitatively (Table 3). Quantitative investigation of DWI, ADC map showed that the re- sults obtained from these parameters were not com- patible with histological findings in 14 cases (31.8%). Sensitivity, specificity, PPV, NPV, and accuracy were reported to be 63.6, 72.7, 70, 66.7, and 68.2%, respec- tively. Considering the ROC curve, the AUC was cal- culated as 0.73, which was statistically significant (P = .003). On this basis, 648 × 10-6 was determined as the cut-off point for differentiation of benign lesions Table 2. Diagnostic indices of T2-W, DCE, DWI, and MRS Sequences sensitivity specificity Positive Predictive Value Negative Predictive Value Precision T2-W 81.8% (18/22) 95.5% (21/22) 94.7% (18/19) 84% (21/25) 88.6% (39/44) DWI 72.7% (16/22) 86.4% (19/22) 84.2% (16/19) 76.0% (19/25) 79.5% (35/44) DCE 59.1% (13/22) 54.5% (12/22) 56.5% (13/23) 57.1% (12/21) 56.8% (25/44) MRS 71.4% (15/21) 90.5% (19/21) 88.2% (15/17) 76.0% (19/25) 81.0% (34/42) Abbreviations: T2-W: T2-weighted; DWI: Diffusion Weighted Imaging; DCE: Dynamic Contrast Enhanced; MRS: Magnetic Reso- nance Spectroscopy Descriptive indices Mean ADC1 Quantity of MRS2 Mean 684.6×10-6 1.14 Standard Deviation 138.9×10-6 1.47 minimum 402×10-6 0.02 maximum 978×10-6 7.16 1: Apparent diffusion coefficient; 2: Magnetic resonance spectros- copy Table 3. Descriptive indices for mean ADC and quantity of MRS MpMRI for the detection of inner gland prostate cancer-Ghane et al. Urological Oncology 554 Vol 16 No 06 November-December2019 555 from malignant ones. The quantitative investigation of MRS showed that the results were not compatible with pathology findings in 11 cases. Sensitivity, specificity, PPV, NPV, and accuracy were reported as 68.2, 81.8, and 78.8, 72, and 75%, respectively. Considering the ROC curve, the AUC was calculated as 0.73 which was statistically significant (P = .004). Based on this result, 0.91 was determined as the cut-off point for differentia- tion of benign lesions from malignant types. Our results showed that there was no significant differ- ence between ADC and MRS in diagnosing the type of tumor (P = .97). The mean ADC and the quantity of MRS were combined together. The results of this combination in diagnosing the type of tumor were not compatible with the pathology findings in 18 cases. Sensitivity, specificity, PPV, NPV, and accuracy were reported to be 59.1%, respectively. The AUC was also estimated as 0.59, which was not statistically significant (P = .23) (Figure 4). DISCUSSION The current study was carried out to evaluate the di- agnostic indices of mpMRI for detection of malignant or benign type of tumor in patients suspicious of inner gland prostate cancer. Our results indicated that T2W had the highest sensitivity and specificity while DCE had the least. The sensitivity and specificity of qualita- tive DWI and MRS techniques were found to be sim- ilar. DWI and MRS were not significantly different compared to T2W. However, DCE specificity had a significant difference compared to T2W. Each of these techniques might report false positive results because of the difficult differentiation between prostate cancer and benign hyperplasia Due to the hypervascularity of BPH in DCE and the low amount of ADC in DWI and ADC maps, there is an overlap between the amounts of ADC in stromal BPH with the amounts in prostate cancer.(4,6,7) In addition, based on the data obtained from DCE imaging, 9 of the 22 cases with malignant lesions had persistent enhancement which might be attributed to the presence of fewer arteries in the tumor.(8) In MRS, the different amount of metabolites in various parts of the prostate (such as the difference in the peri-urethral zone from other zones or the higher citrate concentra- Figure 2. Classification of MRS I: Cho is significantly lower than citrate (<<) II: Cho is elevated but still lower than citrate (<) III: Cho is approximately on the same level as citrate (=) IV: Cho is elevated compared to citrate (>) V: Cho is significantly elevated compared to citrate (>>) Figure 3. ROC curve for comparison of the diagnostic value of the three sequences of DWI, DCE, and MRS with the sequence of T2 in diagnosing the type of tumor Figure 4. ROC curve for comparison of mean ADC, MRS, and combination of mean ADC, MRS MpMRI for the detection of inner gland prostate cancer-Ghane et al. tion in glandular proliferation) as well as the extensive range of metabolites in tumors of the internal portion of prostate may result in false positive results.(9,10) The sen- sitivity and specificity of T2W were the highest which could be due to anatomical characteristics.(11) Nonethe- less, benign lesions such as chronic inflammation of the prostate, atrophy, scar, benign hyperplasia of the pros- tate, post-biopsy bleeding, and the effects of hormone therapy or radiotherapy can mimic the tumor tissue in T2W. In this study, the number of false positive cases in T2W reduced with the use of MRI at appropriate time intervals with respect to previous biopsy and also elimi- nation of patients who had received previous treatment. Our findings also indicated that adding functional tech- niques to T2W does not improve diagnostic indices of the inner gland. Regarding mpMRI, our results are in agreement with the studies by Delongchamps et al.(3) and Hoeks et al.(2) Li et al. (2006) indicated that add- ing DCE to T2W increased the diagnostic precision of prostate cancers within the transition zone.(12) Puech and colleagues investigated the diagnostic value of DCE and the consequences of its elimination from mpMRI. They showed that this technique is able to identify un- detectable lesions on T2W and DWI.(13) However, in the present study, DCE had the lowest diagnostic value and its addition to T2W did not increase diagnostic perfor- mance. Meanwhile, based on the results, there was no lesion identified by DCE which had not been observed on T2W. Hong Li et al. considered monotonous low intensity signal on T2W, homogeneous enhancement in DCE, and irregular margins between the lesion and the central zone on T2W and DCE which could not be easily detected in T2W, sometimes, in favor of cancer. However, in the present study, the diagnostic value of DCE was based on the TIC curve, ascribing to the dif- ferences in methodologies. The results of this study indicated that T2W and DWI techniques were appropriate for detection of type of tu- mor within the transition zone, but DCE did not pro- vide any further information. Meanwhile, gadolinium increases the time and cost of the test.(14-17) In the present study, in addition to the qualitative inves- tigation of DWI, the intensity of the signal was achieved by drawing regions of interest (ROI) on ADC maps of the lesion. This measure shows the degree of diffusion of water molecules. In scientific terms, the diffusion of water molecules is more restricted in malignant le- sions and so, the signal intensity decreases in greater quantities.(2) This quantitative evaluation stated that the mean ADC is useful in differentiating between benign and malignant tumors which is in line with a qualitative study by Schimoller et al (2014).(18) Our findings also revealed that quantitative MRS was able to differentiate between benignancy and malig- nancy. The comparison between mean ADC and MRS quantity indicated that the two techniques were not sig- nificantly different in diagnosing the tumor type. In ad- dition, the combination of mean ADC and MRS quan- tity showed weakness in differentiating benign from malignant lesions. However, more data is needed for a more accurate report. There were some limitations in the present study. Our sample size was restricted due to the fact that the prev- alence of transition zone tumor is much less than pe- ripheral zone tumor which makes it difficult to collect more samples in a limited time. Second, the results of DCE and TIC curve obtained from mean curve software might have been different if another type of software was used. The third limitation was related to the gold standard diagnosis method. mpMRI TRUS fusion bi- opsies improve the detection of clinically significant cancers compared to systematic TRUS-guided biopsies, however, fusion biopsies alone fail to diagnose 8.3% of cancers including 6.7% of significant cancers.(19) CONCLUSIONS The results of this study showed that T2W is the best MRI imaging technique for the diagnosis of type of tu- mor in the inner gland of prostate. Adding functional techniques did not increase the diagnostic value of tu- mor detection in this zone. ACKNOWLEDGEMENT The authors would like to thank the staff of Medical Imaging Center of Payambaran Hospital, Tehran, Iran for their support during data collection of this study. The authors also acknowledge the valuable comments and suggestions of the reviewers, which improved the quality of this paper. CONFLICT OF INTEREST The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publica- tion of this paper. REFERENCES 1. Pakzad R, Rafiemanesh H, Ghoncheh M, et al. Prostate Cancer in Iran: Trends in Incidence and Morphological and Epidemiological Characteristics. Asian pac.j.cancer prev.2016;17:839-43. 2. Hoeks CM, Hambrock T, Yakar D, et al. Transition zone prostate cancer: detection and localization with 3-T multiparametric MR imaging. Radiology. 2013;266:207-17. 3. Delongchamps NB, Rouanne M, Flam T, et al. Multiparametric magnetic resonance imaging for the detection and localization of prostate cancer: combination of T2-weighted, dynamic contrast-enhanced and diffusion-weighted imaging. BJU Int. 2011;107:1411-8. 4. Haider MA, van der Kwast TH, Tanguay J, et al. Combined T2-weighted and diffusion- weighted MRI for localization of prostate cancer. AM J ROENTGENOL. 2007;189:323- 8. 5. Obuchowski NA, Zhou XH. Prospective studies of diagnostic test accuracy when disease prevalence is low. Biostatistics. 2002;3:477-92. 6. Noworolski SM, Vigneron DB, Chen AP, Kurhanewicz J. Dynamic contrast-enhanced MRI and MR diffusion imaging to distinguish between glandular and stromal prostatic tissues. Magn Reson Imaging. 2008;26:1071- 80. 7. Padhani AR, Gapinski CJ, Macvicar DA, et al. Dynamic contrast enhanced MRI of prostate cancer: correlation with morphology and MpMRI for the detection of inner gland prostate cancer-Ghane et al. Urological Oncology 556 Vol 16 No 06 November-December2019 557 tumour stage, histological grade and PSA. Clin Radiol. 2000;55:99-109. 8. Josefsson A, Wikstrom P, Granfors T, et al. Tumor size, vascular density and proliferation as prognostic markers in GS 6 and GS 7 prostate tumors in patients with long follow- up and non-curative treatment. Eur Urol. 2005;48:577-83. 9. Mueller-Lisse UG, Scherr MK. Proton MR spectroscopy of the prostate. Eur J Radiol. 2007;63:351-60. 10. Hoeks CM, Barentsz JO, Hambrock T, et al. Prostate cancer: multiparametric MR imaging for detection, localization, and staging. Radiology. 2011;261:46-66. 11. Akin O, Sala E, Moskowitz CS, et al. Transition zone prostate cancers: features, detection, localization, and staging at endorectal MR imaging. Radiology. 2006;239:784-92. 12. Li H, Sugimura K, Kaji Y, et al. Conventional MRI capabilities in the diagnosis of prostate cancer in the transition zone. AM J ROENTGENOL. 2006;186:729-42. 13. Puech P, Sufana-Iancu A, Renard B, Lemaitre L. Prostate MRI: can we do without DCE sequences in 2013? Diagn Interv Imaging. 2013;94:1299-311. 14. Chesnais AL, Niaf E, Bratan F, et al. Differentiation of transitional zone prostate cancer from benign hyperplasia nodules: evaluation of discriminant criteria at multiparametric MRI. Clin Radiol. 2013;68:e323-30. 15. Scialpi M, Rondoni V, Aisa MC, et al. Is contrast enhancement needed for diagnostic prostate MRI? Transl Androl Urol. 2017;6:499-509. 16. Junker D, Steinkohl F, Fritz V, et al. Comparison of multiparametric and biparametric MRI of the prostate: are gadolinium-based contrast agents needed for routine examinations? World J Urol. 2019;37:691-9. 17. Thestrup KC, Logager V, Baslev I, Moller JM, Hansen RH, Thomsen HS. Biparametric versus multiparametric MRI in the diagnosis of prostate cancer. Acta Radiol Open. 2016;5:2058460116663046. 18. Schimmoller L, Quentin M, Arsov C, et al. MR-sequences for prostate cancer diagnostics: validation based on the PI-RADS scoring system and targeted MR-guided in-bore biopsy. Eur Radiol. 2014;24:2582-9. 19. Kaushal R, Das CJ, Singh P, Dogra PN, Kumar R. Multiparametric magnetic resonance imaging-transrectal ultrasound fusion biopsies increase the rate of cancer detection in populations with a low incidence of prostate cancer. Investig Clin Urol. 2019;60:156-61. MpMRI for the detection of inner gland prostate cancer-Ghane et al.