official author order -...

Official author order: Rui Wang, MD1 Email: [email protected] Jing Wang, PhD2 Email: [email protected] Ge Gao, MD1 Email: [email protected] Juan Hu, MD1 Email: [email protected] Yuanyuan Jiang, MD1 Email: [email protected] Zhenlong Zhao, MD1 Email: [email protected] Xiaodong Zhang, PhD1 Email: [email protected] Yu-Dong Zhang, MD3 Email: [email protected] Xiaoying Wang, MD1 Email: [email protected]

1 Department of Radiology, Peking University First Hospital, Beijing, China, 100032 2 Center for Medical Device Evaluation, CFDA, Beijing, China, 100044 3 Department of Radiology, the First Affiliated Hospital with Nanjing Medical University, Nanjing, China, 210009 Please corresponding to: Yu-Dong Zhang, M.D., Department of Radiology, the First Affiliated Hospital with Nanjing Medical University 300, Guangzhou Road, Nanjing, Jiangsu Province, China, 210009 E-mail: [email protected] Tel: +86-158-0515-1704 Xiaoying Wang, M.D., Department of Radiology, Peking University First Hospital, No. 8, Xishenku St., Xicheng District Beijing, China, 100032 E-mail: [email protected] Tel: +86-135-1107-7396 Running title:

Research. on September 12, 2018. © 2017 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on January 31, 2017; DOI: 10.1158/1078-0432.CCR-16-2884

http://clincancerres.aacrjournals.org/

Image-based predictive nomogram in prostate cancer Relevant keywords: Prostate cancer; PSA; multi-parameter MRI; machine learning analysis; nomogram; support vector machine STATEMENT OF TRANSLATIONAL RELEVANCE 1. This is an prospective study with the largest patient capacity in China sought to answer the question that it is time to consider a role for MRI before prostate biopsy. 2. Featured with high accuracy and low false positive rate, the newly developed nomogram based on pre-biopsy clinical and imaging data can help to predict the clinical outcome accurately in 92.8% patients, even without biopsy results. This finding strongly supports the theory that prostate mp-MRI could be the first investigation of a man with a raised PSA before invasive biopsy, avoiding over-diagnosis and overtreatment.




1

ABSTRACT:

Purpose: To investigate whether pre-biopsy multi-parametric (mp) MRI can help to improve predictive performance in prostate cancer (PCa). Experimental Design: Based on a support vector machine (SVM) analysis, we prospectively modeled clinical data (age, PSA, DRE, TRUS, PSA density and prostate volume) and mp-MRI findings (PI-RADS score and TNM stage) on 985 men to predict the risk of PCa. The new nomogram was validated on 493 patients treated at a same institution. Multivariable Cox regression analyses assessed the association between input variables and risk of PCa. And area under the receiver operating characteristic curve (Az) analyzed the predictive ability. Results: At 5-yr follow-up period, 34.3% of patients had systemic progression of PCa. Nomogram (SVM-MRI) predicting 5-yr PCa rate trained with clinical and mp-MRI data was accurate and discriminating with an externally validated Az of 0.938, positive predictive value (PPV) of 77.4% and negative predictive value of 91.5%. The improvement was significant (p < 0.001) comparing to the nomogram trained with clinical data. When stratified by PSA, SVM-MRI nomogram had high PPV (93.6%) in patients with PSA > 20 ng/ml, with intermediate to low PPV in PSA 10-20 ng/ml (64%), PSA 4-10 ng/ml (55.8%) and PSA 0-4 ng/ml (29%). PI-RADS score (Cox hazard ratio [HR]: 2.112; p < 0.001), PSA level (HR: 1.435; p




2

< 0.001) and age (HR: 1.012; p = 0.043) were independent predictors of PCa. CONCLUSION: Featured with low false positive rate, mp-MRI could be the first investigation of a man with a raised PSA before prostate biopsy. Key words: Prostate cancer; PSA; multi-parametric MRI; machine learning analysis; nomogram; support vector machine

Introduction

Prostate cancer (PCa) is the most commonly diagnosed cancer that affects elderly men and therefore

is a bigger health concern in developed countries (1), and now the incidence is also rapidly increasing

in China (2). The use of prostate-specific antigen (PSA) in combination with digital rectal examination

(DRE) is a widely adopted, population-based screening program in clinical practice that can effectively

detect PCa at earlier, asymptomatic stages (3, 4). However, this traditional screening plan has recently

come in for a lot of criticism that due to its noticeable limitations for cancer detecting, disease

monitoring and patient management (5). For its natural feature of variability, PSA-based screening

lead to an increase in the number of unnecessary biopsies and high risk of overtreatment (6, 7). Now

there is a lack of evidence to support PSA-based screening can influence PCa mortality. And a prostate

lung colorectal & ovarian cancer screening (PLCO) program has been discredited because of poor

survival advantage (8). Therefore, there is an imminent need for simplified predictive tools that




3

include novel pretreatment variables that can extend the clinical performance of traditional

PSA-based nomogram.

Over the past decades, multi-parametric MRI (mp-MRI) was only recommended for prostate cancer

staging after biopsy (9, 10). Recently, it has resulted in great ability in determining tumorous

aggressiveness, and potentially, predicting the risk of disease progression or recurrence (11-14). Many

urologists now request a pelvic MRI scan as the first investigation of a man with a raised PSA before

the prostate biopsy (15, 16). This allows men with a normal scan to be monitored, and if the men with

an abnormal scan, this sensible plan can permit better-informed decisions to be made about

recommendations for prostate biopsy, watchful waiting, radiotherapy or prostatectomy (3, 9, 17).

These findings suggested mp-MRI could do more in entire clinical workflow for patient management.

Therefore, the purpose of this study is to investigate whether pre-biopsy mp-MRI findings, when

converted into a prognostic nomogram, can help to improve predictive performance in a large

contemporary cohort of Chinese patients undergoing PSA screening, with a focus on 5-yr PCa rate.

Materials and Methods:

Patient population and follow-up

This study was approved by our local institutional review board (Peking University First Hospital,

Beijing, China). Between July 2002 and December 2009, 1778 consecutive patients with elevated PSA

undergoing a pre-biopsy prostate mp-MRI were prospectively recruited. Written informed consents

were obtained from all patients. Before MR examination, a questionnaire survey, including the latest




4

PSA level within one month, DRE reports, transrectal ultrasound (TRUS) findings, previous biopsies,

previous prostate MRI scans, history of previous prostate treatment or intervention, was conducted.

Patients were consistently followed with intervals of 6 mo. to 2 years, and were censored at the

occurrence of PCa, emigration, or 31 December 2014, whichever came first. The follow-up included

measurements of serum PSA level, and/or a DRE, and/or MR examination, and/or an invasive

TRUS-based biopsy. At the end of follow-up period, PCa was determined by histopathological

examination, i.e., TRUS-based biopsy, transurethral resection of the prostate (TURP) and radical

prostatectomy (RP). Patients with initial negative biopsy were specially followed by PSA, DRE, MRI,

and/or repeated TRUS-biopsy until the end of follow-up period. Patients without histopathological

results were treated as PCa if they had consistently elevated PSA level (>10 ng/ml after three rounds

of PSA testing), and/or significantly positive findings on imaging examination (i.e., CT, MR or bone

scan). The exclusion criteria included: (a) patient underwent prostate biopsy before mp-MRI (n = 6,

0.33%); (b) patients failed to receive standardized MR or underwent the MR examination from

outside institutions (n = 3, 0.17%); (c) patients with inadequate information that cannot determine

the clinical outcome within 5-yr following MR examination (n = 291, 16.4%). At last, 1478 patients

(median age, 68 yr; age range, 56-81 yr) were eligible for clinical evaluation.

MR examination

All imaging examinations were performed with 1.5-T MR scanners (Signa HDxt, GE Medical System,

Milwaukee, Wis) using pelvic phased-array coils. As per the standard prostatic MRI at our institution,

the protocol included a combination of T2-weighted imaging (T2WI), diffusion-weighted imaging




5

(DWI), MR spectroscopy imaging (MRSI), and dynamic contrast-enhanced (DCE) MR imaging. The MR

parameters and representative MR imaging cases are presented in supplementary data [S-1; S-2].

Imaging analysis

The prostate MR images were retrospectively interpreted by two radiologists with 3 years (R.W.) and

10 years (H.W.) of experience who were blinded to the histologic results and all clinical information.

In each patient, the radiologists first identified the most suspicious cancer lesion (leading lesion) in

the peripheral zone (PZ) and/or the transition zone (TZ). For the patient with multiple lesions, the

largest sized lesion, and/or dominantly low signal intensity (SI) on apparent diffusion coefficient (ADC)

maps, and/or suspected extracapsular extension (ECE), and/or suspected eminal vesicle invasion (SVI),

was defined as the leading lesion (supplementary data [S-3]). The following MR features were then

recorded according to the guidelines of European Society of Urogenital Radiology (ESUR) (9): 1) tumor

location (PZ or TZ); 2) a Prostate Imaging and Reporting and Data System (PI-RADS) score using v2 (18);

3) MR-detected ECE, SVI, local lymph node (LN) and local bone metastasis. During the image

interpretation, any inter-reader disagreement in qualitative assessments was discussed until

consensus reached. This procedure was performed after a statistical measurement of inter-observer

variability.

Model development

The patients were divided into two groups in a randomized fashion: one group was designated the

training group (n=985) for training the predictive model, and the other group was designated the

validation group (n=493) for the evaluation of the accuracy of the predictive model. This randomized




6

fashion for the classification of patients is an advisable way to avoid model overfitting. Patient age (<

60 ys and ≥ 60 ys of age), DRE findings (positive or negative), TRUS results (positive or negative),

prostate volume, PSA level, PSA density (PASD = PSA/prostate volume), and MRI findings including

PI-RADS score, ECE, SVI, local LN and local bone metastasis, were included as input data. The prostate

volume was determined by TRUS or MRI, wherein the volume is the products of Height (H), Width (W)

and Length (L) of prostate gland, as well as with a coefficient of 0.52 (19).

Predictive model was constructed from data of the training group composed of 985 patients by using

a novel support vector machine (SVM) analysis. SVM is a supervised machine learning technique used

for classification and regression analysis. SVM algorithm tries to construct an optimal separating

hyperplane that maximizes the margin, where the margin is the largest distance to the nearest

training data point of any class (20, 21). Unlike the traditional artificial neural networks (ANNs), SVM

does not have to undertake a trial and error parameter decision process, while determines optimal

performance conditions automatically if the kernel type is set. In this study, SVM with radial basis

function (RBF) kernel was applied to resolve the two class problems, clinical progression of PCa or not

(non-PCa). A RBF kernel, K, maps the original data with the kernel function as:

K( ) = exp(-g|| -t||2)

[eq.1],

where and t are two feature vectors, and Gamma (g) controls the shape of the decision hyperplane.

As the direct output value of SVM analysis does not show probabilities of PCa, we converted their

output values to the probabilities (Pi) by applying a sigmoid function as follows:

= [eq.2],




7

where is the output value of SVM analysis. The value of Pi indicates the probabilities that the

patient has PCa.

In the first model (SVM-clinical), six input parameters, i.e. patient age, PSA level, prostate volume,

PSAD, DRE and TRUS findings were used. In the second model (SVM-MRI), the mp-MRI findings

(PI-RADS score, ECE, SVI, local LN and bone metastasis) were added.

Statistical analysis

Regarding the reproducibility of MR measurements, the qualitative results (i.e., PI-RADS score, ECE or

SVI) between the two readers were assessed by kappa (κ) statistics test.

The performance of two predictive models was analyzed from 493 validation data, by using a binary

receiver operating characteristic (ROC) regression analysis and quantified by the areas under the ROC

curves (Az). The predictive sensitivity (SEN), specificity (SPE), positive predictive value (PPV), negative

predictive value (NPV) and overall accuracy (ACC) were calculated at a cutoff point that maximized

the value of the Youden index. In order to determine the independent risk factors who were

associated with the time to progression of PCa, a Kaplan-Meier analysis with log-rank test was firstly

employed to identify the univariate associations between the predictors and progression of PCa.

Factors significant in Kaplan-Meier test were further analyzed by using a multivariable Cox regression

analysis to determine the independent risk factors who were associated with the time to progression

of PCa. Statistical analysis was performed with commercially available software (SPSS 15.0 for

Windows; SPSS, Chicago, IL, USA). A p-value of < 0.05 was considered statistically significant.




8

Results Table 1 lists clinical and MR imaging characteristics of the training and validation cohorts. As of Dec

2014, overall 507 (34.3%) of the 1478 patients in the study had a 5-year clinical progression of PCa.

The TRUS-biopsy was performed initially in 429 patients and produced a 79.0% (339/429) positive

detection rate. The remaining 168 patients with PCa was detected by repeated biopsy (n = 62), and

via comprehensive clinical-imaging analysis (n = 101), and via TURP incidentally (n = 5). The

cumulative rate of PCa was 29.6% (437/1478) at 1-yr, 31.8% (470) at 3-yr and 34.3% (507) at 5-yr,

respectively. Among of all positive cases, 47.1% (239/507) patients underwent endocrine therapy,

13.2% (67/507) underwent prostate androgen deprivation therapy, 18.3% (93/507) underwent active

RP, 15.2% (77/507) underwent radiation therapy, and 2.2% (11/507) underwent chemotherapy.

During the 5-yr follow-up period, 7.1% (36/507) patients died of PCa and 12.8% (65/507) died of

another cause.

Supplementary data [S-4] shows the workflow of machine learning analysis with SVM for model

development in 985 training data. Figure 1 shows the performance of SVM-clinical and SVM-MRI

model in training data and validation data, respectively. Using traditional clinical input parameters,

i.e., patient age, PSA level, prostate volume, PSAD, DRE and TRUS findings, the Az of SVM-clinical

model for PCa prediction was 0.716 (95% confidence interval [CI]: 0.686-0.744) for training data and

0.715 (95% CI: 0.673-0.754) for validation data. With addition of mp-MRI findings, the Az (training:

0.930, 95%CI: 0.912-0.945; validation: 0.932, 95%CI: 0.906-0.953) of SVM-MRI model becomes

significantly larger (p < 0.001).

Table 2 shows the internal validity of two predictive models, i.e., SVM-clinical vs. SVM-MRI, to predict

5-year PCa rate for different cutoff points in the 985 modeling and 493 validation data, respectively.

Figure 2 shows 1-yr, 3-yr and 5-yr prediction rate functions (PRFs) of PCa estimated with SVM-clinical

and SVM-MRI model in 493 validation data. The cumulative predictive accuracy estimated by

SVM-MRI model was 92.8% at 1-yr, 87.9% at 3-yr, and 84.6% at 5-yr, respectively, significantly higher




9

than those estimated by SVM-clinical model (68.9%, 67.9% and 68.6%, respectively). It shows that the

PRF constructed by SVM-MRI model is excellently close to patients’ actuarial PRF until 32 mo after the

follow-up. While both two models have increased bias with long-term follow-up period. When

stratified by PSA level, the SVM-MRI nomogram had high positive predictive value (PPV; 93.6%) in

patients with PSA > 20 ng/ml, while with intermediate to low PPVs in PSA 10-20 ng/ml (64%), PSA

4-10 ng/ml (55.8%) and PSA 0-4 ng/ml (29%). The SVM-MRI model had a NPV of higher than 90% in all

four groups with different PSA level (Figure 3).

Unadjusted and multivariable adjusted hazard ratios (HRs) with cox proportional hazards model are

presented in supplementary data [S-5]. In univariate analyses, age, baseline PSA level, prostate

volume, PI-RADS score were all significantly associated with an increased risk of PCa. In adjusted

analysis, PI-RADS score had the highest Cox HR (2.112; p < 0.001) for prediction of time to PCa,

followed by PSA level (HR: 1.435; p < 0.001), and age (HR: 1.012; p = 0.043). The distribution of

PI-RADS scores predicting the probability of PCa is shown in Table 3, showing the incidence of PCa

increased with the mp-MRI score. The vast majority of patients (512/1478, 34.6%) were given a score

of 4 and 5, signifying an mp-MRI report of an organ-confined tumor. The incidence of PCa was 2.9% in

whom no tumor was apparent (score 1-2) and 3.5% in whom PCa could not be excluded (scores 3).

Using a cutoff value of score 3, PI-RADS produced good predictive ability in PCa, with a SEN of 81.3%,

SPE of 89.7%, PPV of 80.5% and NPV of 90.2%, respectively. Unilateral or bilateral ECE was reported in

286/1478 (19.4%) patients, SVI was reported in 137/1478 (9.3%) patients, local LN metastasis was

reported in 73/1478 (4.9%) patients, and local bone metastasis was reported in 113/1478 (7.6%)

patients.

Discussion

To our knowledge, this is the first prospective study to investigate the predictive role of pre-biopsy

mp-MRI for PCa with a long-term follow-up in 1478 consecutive Chinese patients. We report that, at

5-yr follow-up period, 34.3% of patients had systemic progression of PCa, and 29.6% was detected

within initial 1-yr follow-up, while only 2.5% was detected within the following 4-yr period. In addition,




10

we noted that the machine learning-based model using the input variables of clinical variables and

mp-MRI findings (SVM-MRI) shows high sensitivity (83.5%) and specificity (87.8%) to predict PCa

statistically more accurate (86.4%) than PSA-based screening program. And importantly, high NPVs

(larger than 90% in any PSA level) made the mp-MRI extraordinarily useful for initial evaluation before

an invasive biopsy.

The conventional diagnostic tools for detection of PCa are DRE, PSA, and TRUS-guided biopsies (3),

while mp-MRI is commonly recommended for cancer staging after biopsies. However, in about 18% of

all patients, PCa is detected by a suspect DRE alone, irrespective of the PSA level (22). A suspect DRE

in patients with a PSA level of up to 2 ng/ml has a PPV of 5-30% (23, 24). In our study, only 9.5% of all

patients was detected positively by DRE alone, and 7.0% by TRUS alone. A suspect DRE or TRUS in

patients with a PSA level of up to 4 ng/ml has a PPV of 36.7%. To address these limitations, we

proposed a prognostic nomogram to more accurately quantify the individual probability of PCa based

a machine learning approach. We noted that, even loading an advanced SVM analysis, the model

constructed from clinical input variables only increased the PPV to 40.5%, and overall performance (Az)

was 0.715. When mp-MRI was added, the Az increased to 0.938, and overall predictive accuracy

increased to 86.4%. Additionally, the new nomogram increased NPVs from 71.3% to 91.5%. By this

improvement, 20.2% patients with PSA testing positive could avoid undergoing invasive biopsy. This

result confirms strongly the thesis that mp-MR findings can complement the standard clinical

nomogram for the improvement of the predictive performance. The nomogram by integrating

mp-MRI and clinical variables could be noninvasive and prospective, thus allows for immediate

identification of the patients accounted for biopsy or not. This thesis now receives more support in

the urologic literatures (25-27).

The significant contribution of mp-MRI to predict PCa may be explained by the assumption that

PI-RADS score serves as a possible indicator for the probability of PCa (28-30). We noted that PI-RADS

score > 3 resulted in only 9.8% false negative results. And among all mp-MRI parameters, PI-RADS

score was the only independent predictor of detection rate of PCa, as well as with the highest Cox HR

value among all independent predictors. However, mp-MRI has relatively low ability in detection of




11

clinically insignificant PCa. As most of insignificant tumors have volume less than 0.5 ml, or PSA level

less than 10 ng/ml, or Gleason score less than 3+3, which are commonly invisible on mp-MR images.

We observed that, among 816 patients who had a PI-RADS score 1 and score 2 (MR-invisible), the rate

of PCa was 5.3% (43/816). Among patients who had a PSA level less than 4 ng/ml and a PSA level

between 4-10 ng/ml, the rate of PCa was 6.4% (12/188) and 19.4% (104/535), respectively. The PPV of

SVM-MRI model for predicting PCa in this two groups was only 29.0% and 55.8%, respectively. While

among 351 patients who had PSA larger than 20 ng/ml, the PPV increased to 93.6%. However, we

noted The SVM-MRI model had high NPVs (> 90%) regardless of PSA level. This finding strongly

supports that mp-MRI can help to detect the most biologically significant PCa and predict all negative

cases, thus allows to immediately identify the patients who are accounted for active RP treatment and

who only need an active surveillance.

We noted that among all studied patients, the first 1-year true positive rate is 29.6% (n = 437), and the

first 1-year cumulative predictive accuracy by SVM-MRI model is as high as 92.8%. This indicates the

outcome of most patients underwent PSA screening program could be accurately predicted in the first

1-yr follow-up period if they received a pre-biopsy mp-MRI examination, even without invasive TRUS

biopsy. And second, we noted the true positive rate is only 2.5% (n = 70) in the following 4-yr

follow-up period. This indicates that there is limited clinical significance for a long-term follow-up in

patients who are detected negatively in the first 1-yr period.

A couple of limitations of our study warrant mention. First, mp-MRI data were obtained at a 1.5 T MR

scanner with utilization of a surface coil. While at present, MRI at 3 T with quantitative parameters,

i.e., ADCs, Ktrans and texture variables, is often considered state of the art. Second, because of so

long-term periods, the imaging criteria for PCa has been greatly changed (i.e., from early

experience-based norms to a structured PI-RADS v1 and recently updated PI-RADS v2), the decision on

which may make influence on patient management. Moreover, the new PI-RADS v2, even showing

great advantage, still needs to be validated by numerous multicenter studies in the future. Third,

because we did not have a device that supports the TRUS/MR fusion-guided in-bore biopsy, the

histopathology validation is not with a targeted approach, which may disqualify the MRI read outs.




12

Last, this is a single center study, an external validation cohort should be included to test the

reproducibility of established model in future.

Conclusions

We first investigated the systemic outcome of a crowd underwent PSA-based screening and pre-biopsy

mp-MRI, and demonstrated the predictive role of pre-biopsy mp-MRI for PCa by using an advanced

machine learning-based approach. Here we answer two important questions at the beginning of the

paper: (1) Machine learning analysis of mp-MRI findings can help to improve the predictive

performance in PCa, by which, the outcome of 92.8% patients could be accurately predicted in the

first 1-yr follow-up period by combining clinical and mp-MRI findings. (2) Advanced in high PPVs and

NPVs, prostate mp-MRI could be the first investigation of a man with a raised PSA before invasive

biopsy, avoiding over-diagnosis and overtreatment.

References: 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA: a cancer journal for clinicians. 2016;66:7-30. 2. Chen W, Zheng R, Baade PD, Zhang S, Zeng H, Bray F, et al. Cancer statistics in China, 2015. CA: a cancer journal for clinicians. 2016;66:115-32. 3. Mottet N, Bellmunt J, Bolla M, Briers E, Cumberbatch MG, De Santis M, et al. EAU-ESTRO-SIOG Guidelines on Prostate Cancer. Part 1: Screening, Diagnosis, and Local Treatment with Curative Intent. European urology. 2016 Aug 25. pii: S0302 - 2838 (16) 30470-5. DOI: 10.1016/j.eururo.2016.08.003. [Epub ahead of print]. 4. Hayes JH, Barry MJ. Screening for prostate cancer with the prostate-specific antigen test: a review of current evidence. Jama. 2014;311:1143-9. 5. Chou R, Croswell JM, Dana T, Bougatsos C, Blazina I, Fu R, et al. Screening for prostate cancer: a review of the evidence for the U.S. Preventive Services Task Force. Annals of internal medicine. 2011;155:762-71. 6. Schroder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V, et al. Screening and prostate-cancer mortality in a randomized European study. The New England journal of medicine. 2009;360:1320-8. 7. Zhu X, Albertsen PC, Andriole GL, Roobol MJ, Schroder FH, Vickers AJ. Risk-based prostate cancer screening. European urology. 2012;61:652-61. 8. Andriole GL, Crawford ED, Grubb RL, 3rd, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. The New England journal of medicine. 2009;360:1310-9. 9. Barentsz JO, Richenberg J, Clements R, Choyke P, Verma S, Villeirs G, et al. ESUR prostate MR guidelines 2012. European radiology. 2012;22:746-57.




13

10. Heidenreich A, Bastian PJ, Bellmunt J, Bolla M, Joniau S, van der Kwast T, et al. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. European urology. 2014;65:124-37. 11. Yakar D, Debats OA, Bomers JG, Schouten MG, Vos PC, van Lin E, et al. Predictive value of MRI in the localization, staging, volume estimation, assessment of aggressiveness, and guidance of radiotherapy and biopsies in prostate cancer. Journal of magnetic resonance imaging : JMRI. 2012;35:20-31. 12. Jung SI, Donati OF, Vargas HA, Goldman D, Hricak H, Akin O. Transition zone prostate cancer: incremental value of diffusion-weighted endorectal MR imaging in tumor detection and assessment of aggressiveness. Radiology. 2013;269:493-503. 13. Vargas HA, Akin O, Franiel T, Mazaheri Y, Zheng J, Moskowitz C, et al. Diffusion-weighted endorectal MR imaging at 3 T for prostate cancer: tumor detection and assessment of aggressiveness. Radiology. 2011;259:775-84. 14. Schoots IG, Petrides N, Giganti F, Bokhorst LP, Rannikko A, Klotz L, et al. Magnetic resonance imaging in active surveillance of prostate cancer: a systematic review. European urology. 2015;67:627-36. 15. Numao N, Yoshida S, Komai Y, Ishii C, Kagawa M, Kijima T, et al. Usefulness of pre-biopsy multiparametric magnetic resonance imaging and clinical variables to reduce initial prostate biopsy in men with suspected clinically localized prostate cancer. The Journal of urology. 2013;190:502-8. 16. Patel U, Dasgupta P, Challacombe B, Cahill D, Brown C, Patel R, et al. Pre-biopsy 3-Tesla MRI and targeted biopsy of the index prostate cancer: correlation with robot-assisted radical prostatectomy. BJU international. 2017;119:82-90. 17. Cornford P, Bellmunt J, Bolla M, Briers E, De Santis M, Gross T, et al. EAU-ESTRO-SIOG Guidelines on Prostate Cancer. Part II: Treatment of Relapsing, Metastatic, and Castration-Resistant Prostate Cancer. European urology. 2016 Aug 31. pii: S0302 - 2838 (16) 30469-9. DOI: 10.1016/j.eururo.2016.08.002. [Epub ahead of print]. 18. Weinreb JC, Barentsz JO, Choyke PL, Cornud F, Haider MA, Macura KJ, et al. PI-RADS Prostate Imaging - Reporting and Data System: 2015, Version 2. European urology. 2016;69:16-40. 19. Giubilei G, Ponchietti R, Biscioni S, Fanfani A, Ciatto S, F DIL, et al. Accuracy of prostate volume measurements using transrectal multiplanar three-dimensional sonography. International journal of urology : official journal of the Japanese Urological Association. 2005;12:936-8. 20. Kim S, Yu Z, Kil RM, Lee M. Deep learning of support vector machines with class probability output networks. Neural networks : the official journal of the International Neural Network Society. 2015;64:19-28. 21. Balabin RM, Lomakina EI. Support vector machine regression (LS-SVM)--an alternative to artificial neural networks (ANNs) for the analysis of quantum chemistry data? Physical chemistry chemical physics : PCCP. 2011;13:11710-8. 22. Carvalhal GF, Smith DS, Mager DE, Ramos C, Catalona WJ. Digital rectal examination for detecting prostate cancer at prostate specific antigen levels of 4 ng./ml. or less. The Journal of urology. 1999;161:835-9. 23. Loeb S, Catalona WJ. What is the role of digital rectal examination in men undergoing serial screening of serum PSA levels? Nature clinical practice Urology. 2009;6:68-9. 24. Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V, et al. EAU guidelines on prostate cancer. Part 1: screening, diagnosis, and treatment of clinically localised disease. European




14

urology. 2011;59:61-71. 25. Siddiqui MM, Rais-Bahrami S, Turkbey B, George AK, Rothwax J, Shakir N, et al. Comparison of MR/ultrasound fusion-guided biopsy with ultrasound-guided biopsy for the diagnosis of prostate cancer. Jama. 2015;313:390-7. 26. Ahmed HU, Kirkham A, Arya M, Illing R, Freeman A, Allen C, et al. Is it time to consider a role for MRI before prostate biopsy? Nature reviews Clinical oncology. 2009;6:197-206. 27. Moore CM, Robertson NL, Arsanious N, Middleton T, Villers A, Klotz L, et al. Image-guided prostate biopsy using magnetic resonance imaging-derived targets: a systematic review. European urology. 2013;63:125-40. 28. Grey AD, Chana MS, Popert R, Wolfe K, Liyanage SH, Acher PL. Diagnostic accuracy of magnetic resonance imaging (MRI) prostate imaging reporting and data system (PI-RADS) scoring in a transperineal prostate biopsy setting. BJU international. 2015;115:728-35. 29. Hamoen EH, de Rooij M, Witjes JA, Barentsz JO, Rovers MM. Use of the Prostate Imaging Reporting and Data System (PI-RADS) for Prostate Cancer Detection with Multiparametric Magnetic Resonance Imaging: A Diagnostic Meta-analysis. European urology. 2015;67:1112-21. 30. Park SY, Jung DC, Oh YT, Cho NH, Choi YD, Rha KH, et al. Prostate Cancer: PI-RADS Version 2 Helps Preoperatively Predict Clinically Significant Cancers. Radiology. 2016;280:108-16.




15

Tables and legends: Table 1: Clinical and MR imaging characteristics of patients in training and validation cohorts

Variable Training data (n = 985)

Validation data (n = 493) p

Clinical characteristics Median age (IQR) 70 (65-75) 70 (65-75) 0.967 Median follow-up (IQR) 40 (1-80) 38 (1-78) 0.936 Median ng/ml PSA (IQR) 10.3 (6.2-19.1) 9.8 (6.2-18.5) 0.981 Median prostate volume (IQR) 54.5 (35.6-73.1) 51.9 (32.5-63.7) 0.463 Median PSAD 0.19 (0.10-0.40) 0.20 (0.11-0.44) 0.897 DRE positive, n (%) 82 (8.3%) 58 (11.8%) 0.033 TRUS positive, n (%) 73 (7.4%) 30 (6.1%) 0.387 Initial biopsy positive, n (%) 232 (42.6%) 107 (48.0%) 0.175

Diagnosis of PCa Total detecting rate, n (%) 343 (34.8%) 164 (33.3%) 0.255 Detected by histopathology, n (%) 294 (29.8%) 107 (21.7%) Diagnosed by clinical analysis, n (%) 49 (5.0%) 57 (11.6)

MR findings Median PI-RADS score (IQR) 2 (1-4) 2 (1-4) 0.721 ECE positive, n (%) 188 (19.1%) 98 (19.9%) 0.727 SVI positive, n (%) 117 (11.9%) 20 (4.1%) 0.001 Local LN invasion, n (%) 49 (5.0%) 24 (4.9%) 0.929 Local bone metastasis, n (%) 84 (8.5%) 29 (5.9%) 0.073 Final Gleason score

≤ 3+3, n (%) 42 (14.3%) 10 (9.3%) 0.415 3+4, n (%) 56 (19.0%) 23 (21.5%) 0.876 4+3, n (%) 26 (8.8%) 10 (9.3%) 0.924 ≥ 4+4, n (%) 170 (57.8%) 64 (59.8%) 0.931

Treatment* and death rate ET, n (%) 183 (18.6%) 56 (11.3%) ADT, n (%) 54 (5.5%) 13 (2.6%) RP, n (%) 65 (6.6%) 27 (5.5%) RT, n (%) 50 (5.1%) 27 (5.5%) Chemotherapy, n (%) 5 (0.5%) 6 (1.2%) Total death rate, n (%) 73 (7.4%) 28 (5.6%) 0.087 Death by PCa, n (%) 25 (2.5%) 11 (2.2%) 0.488

Note. IQR = interquartile-range, DRE = digital rectal examination, PSA = prostate-specific antigen, PSAD = PSA density, TRUS = transrectal ultrasound, PCa = prostate cancer, PI-RADS = Prostate Imaging and Reporting and Data System, ECE = extracapsular extension; SVI = seminal vesicle invasion, LN = local lymph node (LN), ET = endocrine therapy, ADT = androgen deprivation therapy, RP = radical prostatectomy, RT = radiation therapy. *n is the therapy frequency (one patient may receive more than one therapy).




16

Table 2: Internal validity of the SVM-clinical/SVM-MRI model to predict 5-yr PCa rate for different cutoff points in the validation data set (n = 493)

Cutoff level Performance parameter for SVM-clinical/SVM-MRI modelSEN (%) SPE (%) PPV (%) NPV (%) ACC (%)

> 0.1 97.0/95.7 6.1/50.8 33.9/49.2 80.0/95.9 36.3/65.7 > 0.3 84.8/93.3 37.9/78.7 40.5/68.6 83.3/95.9 53.5/83.6 > 0.5* 27.4/83.5 89.9/87.8 57.7/77.4 71.3/91.5 69.2/86.4 > 0.7 4.3/56.7 99.4/99.4 77.8/97.9 67.6/82.2 67.7/85.2

Note. -the cutoff level is the level of predictive probability (Pi). There are the points (the cutoff levels) over the receiver operator curve. For example, accuracy of 80% for a cutoff level of 0.5 means that the accuracy of the predictive model with a 50% probability or more for prediction of PCa is 80%. *using the cutoff value of 0.5, SVM-MRI has significantly (p < 0.001) higher SEN, PPV, NPV and ACC than SVM-clinical model. Table 3: Distribution of the 1478 patients according to PI-RADS score, with actual recorded events of PCa for each score PI-RADS Distribution, N (%) Actual PCa, N (%) SEN, % SPE, % PPV, % NPV, % Cutoff

score 1 576 (39.0%) 30 (2.0%) 94.1% 56.2% 52.9% 94.8% > 1 score 2 240 (16.2%) 13 (0.9%) 91.5% 79.6% 70.1% 94.7% > 2 score 3 150 (10.1%) 52 (3.5%) 81.3% 89.7% 80.5% 90.2% > 3 score 4 222 (15.0%) 129 (8.7%) 55.8% 99.3% 97.6% 81.1% > 4 score 5 290 (19.6%) 283 (19.1%)

Note. SEN = sensitivity, SPE= specificity, PPV = positive predictive value, NPV = negative predictive value. Figure 1: the performance of two predictive nomograms, i.e., SVM-clinical vs. SVM-MRI, in prediction

of PCa in training data (a) and validation data (b). Figure 2: 1-yr (a), 3-yr (b) and 5-yr (c) prediction rate functions (PRFs) of PCa by SVM-clinical and

SVM-MRI model in 493 validation data. It shows that SVM-MRI is dominantly superior to SVM-clinical

model in determination of patients’ short-term outcome (less than 32 mo, red arrow). Figure 3: the predictive performance of SVM-MRI model in four groups stratified by PSA level. PPV =

positive predictive value; NPV = negative predictive value.




SVM-clinicalSVM-MRI

Training data (n = 985)

Az = 0.716

Az = 0.930

A

Figure 1




SVM-clinicalSVM-MRI

Vadilation data (n = 493)

Az = 0.715

Az = 0.932

B

Figure 1




actuarial PRFPRFby SVM-clinicalPRFby SVM-MRI

1-Year Progression Function

One

Min

us C

um D

etec

tion

Rat

e

Surv (mo)

ACC = 92.8%

ACC = 68.9%

N = 167

A

Figure 2




Surv (mo)


One

Min

us C

um D

etec

tion

Rat

e actuarial PRFPRFby SVM-clinicalPRFby SVM-MRI

ACC = 67.9%

ACC = 87.9%

N = 290

B

Figure 2




actuarial PRFPRFby SVM-clinicalPRFby SVM-MRI

One

Min

us C

um D

etec

tion

Rat

e


Surv (mo)

ACC = 84.6%

ACC = 68.6%

N = 370

C

Figure 2




0%

20%

40%

60%

80%

100%

120%

0-4 ng/ml 4-10 ng/ml 10-20 ng/ml > 20 ng/ml

PPVs NPVs

29%

55.8%

64%

93.6%98.

1%

94.3%

90.5% 91.

2%

PSA scale

Pred

ictiv

e va

lues

(%)

Figure 3




Published OnlineFirst January 31, 2017.Clin Cancer Res Rui Wang, Jing Wang, Ge Gao, et al. consecutive patientsperformance in prostate cancer: a prospective study in 1478 Pre-biopsy mp-MRI can help to improve the predictive

Updated version

10.1158/1078-0432.CCR-16-2884doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2017/01/31/1078-0432.CCR-16-2884.DC1

Access the most recent supplemental material at:

Manuscript

Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/early/2017/01/31/1078-0432.CCR-16-2884To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-16-2884

http://clincancerres.aacrjournals.org/content/suppl/2017/01/31/1078-0432.CCR-16-2884.DC1

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/early/2017/01/31/1078-0432.CCR-16-2884


official author order -...

Documents