guiqin Liu1, shuaishuai XU1, yongming Dai2, Guangyu WU1, and jianrong XU1
1Radiology, Renji Hospital,Shanghai Jiaotong University School of Medicine, Shanghai, China, 2United Imaing Healthcare, Shanghai, China
Synopsis
To investigate the value of radiomics features from
diffusion-weighted imaging (DWI) in differentiating muscle-invasive bladder
cancer (MIBC) from non-muscle-invasive bladder cancer (NMIBC).
Combining
DWI radiomics features with TUR could improve the sensitivity and accuracy in
discriminating the presence of muscle invasion in bladder cancer for clinical
practice. Multi-center, prospective studies are needed to confirm our results.
INTRODUCTION
To
investigate the value of radiomics features from diffusion-weighted imaging
(DWI) in differentiating muscle-invasive bladder cancer (MIBC) from
non-muscle-invasive bladder cancer (NMIBC).METHODS
This retrospective study included 218
pathologically-confirmed bladder cancer patients (training set: 131 patients,
86 MIBC; validation set: 87 patients, 55 MIBC) who underwent DWI before biopsy
through transurethral resection (TUR) between July 2014
and December 2018. Radiomics models
based on DWI for discriminating state of muscle-invasive were built using
random forest (RF) and all-relevant (AR) methods on the training set and were
tested on validation set. Combination models based on TUR data were also built.
Discrimination performances were evaluated with the area under the receiver
operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity,
F1 and F2 scores. Qualitative MRI evaluation based on morphology was performed
for comparison.RESULTS
Patient population
245 patients were included. After excluding 37
patients in whom radiomics features could not be extracted due to the small
volume of lesions or the limited visibility of images, 218 (169 males; mean
age, 66.1 yrs [range, 37-93]; 141 muscle-invasive tumors) were left for further
analyses. In this patient group, TUR only confirmed 87 muscle-invasive tumors,
and 38.3% (54/141) of RC-confirmed muscle-invasive tumors were misdiagnosed as
non-muscle-invasive tumors at TUR.
Patients
were randomly divided into training set (131 patients; 104 males; mean age,
65.8 yrs [range, 38-86]; 86 muscle-invasive tumors) and validation set (87
patients; 65 males; mean age, 66.5 yrs [range, 37-93]; 55 muscle-invasive
tumors)(Fig 1). No significant difference was observed in age (p=0.696,
wilcoxon rank sum test), gender (p=0.519, chi-squared test) or muscle
invasion (p=0.824, chi-squared test) between the two sets . 73 features with ICC of more than 0.85 were
extracted by different methods, including first order, shape, GLCM, GLRLM, GLSZM,
and NGTDM features. After Boruta selection, 21 all-relevant features were
obtained (Fig 2-3). Internal validation showed no significant difference in AUC
(0.907 vs 0.904, p=0.673, Delong’s test), ACC (0.839 vs 0.816, p=0.480,
McNemar’s test), SEN (0.873 vs 0.855, p=1.000), or SPE (0.781 vs 0.750, p=1.000)
between RandomForest model and All-relevant model for discriminating
muscle-invasive BC (Fig 4).
RandomForest
model was more sensitive than TUR (0.873 vs 0.655, p=0.019, McNemar’s test),
and MRI (0.873 vs 0.764, p=0.181) for discriminating MIBC, but the
difference did not reach statistical significance. When combining the RandomForest
model with TUR, the sensitivity increased to 0.964, significantly higher than
TUR (0.655,
p<0.001), MRI (0.764, p=0.006), and the combination of
TUR and MRI (0.836, p=0.046). Notably, the combination model (RandomForest
model and TUR) had the highest accuracy of 0.897 and F2 score of 0.946 for
discriminating MIBC.DISCUSSION
In this
study, 38.3% (54/141) of RC-confirmed muscle-invasive tumors were misdiagnosed
as non-muscle-invasive tumors at TUR, which is consistent with previous reports
[1,2-4]. Many
reasons account for the poor sensitivity of TUR for discriminating muscle-invasive
tumors, such as sampling error due to incompleteness of TUR, delay in the
interval from TUR to RC, and poor sensitivity of preoperative staging tools [1,2]. Besides, qualitative MRI evaluation only showed a good
inter-observer repeatability (Kappa value = 0.605) and a poor sensitivity
comparable to that of TUR (0.764 vs 0.873, p=0.181).
Our radiomics model exhibited favorable discrimination performance
in internal validation, with an AUC of 0.907 on the test set. The obvious
advantage of TUR is its specificity of 100%, as muscle invasion is confirmed
once observed at TUR specimen without considering the pathological result at
RC. But for detecting highly malignant muscle-invasive BC, what physicians most
importantly need is a more sensitive staging tool with a false negative rate as
low as possible altogether with a relatively high positive predictive value
(PPV). Recall (sensitivity) is more important than precision (PPV). Considering
that F1 score is the harmonic average of the precision and recall, and that F2
score weighs recall higher than precision by placing more emphasis on false
negatives, our radiomics model and combination model showed improved
performance for discriminating muscle-invasive BC compared with TUR and
qualitative MRI evaluation as seen on Table 3.
Another major finding of this study was that a small subset of
all-relevant radiomics features selected by Boruta exhibited an equivalent
performance compared to that of all the extracted features, although the
classification performance using the selected optimal feature subset
outperformed that using the candidate feature set in a previous report [5]. Our finding suggested that radiomics data contained redundant or
irrelevant features and that feature selection should be performed in building
radiomics models.CONCLUSION
Combining DWI radiomics features with TUR could
improve the sensitivity and accuracy in discriminating the presence of muscle
invasion in bladder cancer for clinical practice. Multi-center, prospective
studies are needed to confirm our results.Acknowledgements
The authors thank their colleagues of the department
of radiology of their institute.
References
[1]
Babjuk M, Bohle A, Burger M et al (2017) EAU guidelines on non-muscle invasive
urothelial carcinoma of the bladder: update 2016. Eur Urol 71:447-61
[2]
Karakiewicz PI, Shariat SF, Palapattu GS et al (2006) Precystectomy nomogram
for prediction of advanced bladder cancer stage. Eur Urol 50:1254-1260
[3]
Shariat SF, Palapattu GS, Karakiewicz PI et al (2007) Discrepancy between
clinical and pathologic stage: impact on prognosis after radical cystectomy.
Eur Urol 51:137-149;discussion 49-51
[4]
Svatek RS, Shariat SF, Novara G et al (2011) Discrepancy between clinical and
pathological stage: external validation of the impact on prognosis in an
international radical cystectomy cohort. BJU Int 107:898-904
[5]
Zhang X, Xu X, Tian Q et al (2017) Radiomics assessment of bladder cancer grade
using texture features from diffusion-weighted imaging. J Magn Reson Imaging
6:1281-1288