Patient characteristics
This study was a prospective study of 23 patients (13 males and 10 females) who were treated at Radiation Oncology Department of Cancer Institute, Tehran University of Medical Sciences from October 2016 to April 2018. This study was approved by the Ethics Committee of Tehran University of Medical Sciences (Trial No. 95-04-227-33428, Ethical approval No. IR.TUMS.VCR.REC.1395.17.16) and Iran University of Medical Sciences (Ethics approval No. K17-137) and all patients signed an informed consent. Inclusion criteria were location of tumor about 15 cm above anal verge, tumor influence to perirectal fat (cT3–4) or lymph node involvement, age ≤ 80 years, World health organization (WHO) performance situation of 0–2, normal function of liver and renal system with normal complete blood count (CBC) test, without any previous treatment for the disease. All patient received concurrent nCRT. They received 45–46 Gy external beam radiation in 23–25 fractions with 18 MV photons to the tumor and locoregional disease including pre-sacral and internal iliac lymph nodes with a boost to the tumor for a total of 50–50.4 Gy. Also, concurrent capecitabine at 825 mg/m2 twice daily was used.
Image acquisition
Images were acquired using a Biograph 6 PET/CT scanner (Siemens Medical Solutions). Fields of view were 16 cm and 58 cm in axial and transverse directions, respectively. The CT scan system was a spiral 6 slice scanner with a 50 cm axial field of view. All patients were fasted at least 60 minutes before PET/CT scan with the blood glucose level less than 150 mg/dL. About 60 minutes after the injection of 18F-FDG (5 MBq/kg per body weight), the patients were placed in the scanner and low-dose CT was performed from the base to the mid-thigh of the skull. The PET scan was acquired over the same body area. The CT data reconstructed with 256.256 matrix size and 5-mm slice thickness. The PET data were reconstructed with 128 · 128 matrix size and 5-mm slice thickness.
Tumor segmentation
For each primary cancer site, the three-dimensional gross tumor volume was drawn with two readers: a ten years’ experience radiation oncologist, and a fifteen years’ experience radiologist using a designated multi-platform, free and open-source software package for visualization and medical image computing (3D slicer, version 4.8.1; available at: http://slicer.org/).
Preprocessing and texture feature extraction
Before feature extraction and in order to noise reduction, intensity normalization and discretization, all PET/CT images were pre-processed by the method proposed by Collewet et al. and also discretization to 64 Gy level. In the Collewet et al. method, all image intensities are normalized between µ ± 3σ, where µ is the mean value of gray-levels inside the region of interest (ROI), and σ is the standard deviation [18, 19].
Also, to test the filter effect on radiomic model performance, we applied our feature extraction on PET/CT images with and without processing filters. The filters included LoG filter with sigma 0.5, 1, and 1.5. For feature extraction, we used the freely available radiomic software, imaging biomarker explorer (IBEX) that runs in Matlab platform.
Various radiomic features from different feature sets, including intensity, shape and texture-based features were extracted from processed and un-processed PET/CT images. Extracted features included shape features (n=17), intensity histogram features (n=9), intensity direct (n=19), neighbor intensity difference (n=5), co-occurrence matrix features (COM) (n=19), and gray level run-length matrix features (GLRLM; n=11) [9, 20].
Response assessment
For all patients, surgery was done 6-8 weeks after nCRT. After inking, the specimens were fixed in formalin for 24 hours. The whole tumor and mesorectum were serially sliced, axially, at 3mm intervals, and treatment response was assessed according to the American Joint Committee on Cancer and College of American Pathologists (AJCC/CAP). The Dworak tumor regression grade (TRG) according to AJCC/CAP was established as follows: grade 0 [Pathologic Complete Response (PCR)], which is defined as no viable cancer cell; grade 1 (moderate response), representing single cells or small groups of cancer cells; grade 2 (minimal response), showing residual cancer outgrown by fibrosis; grade 3 (poor response), representing fibrosis outgrown by residual cancer [21–23].
Univariate radiomic analysis
For univariate analysis, significant radiomic features correlating with response were selected and a logistic regression model was used to find their predictive performance, which was based on Area Under Curve (AUC). Also, these features were compared between responder and non-responder groups. A paired t-test was performed to assess the significance of the differences between two groups. Statistical significance was assumed if a two-sided P value< 0.05. All of analysis were done using MedCalc statistical software.