Patient characteristics
This study eligibled 1,084 patients from four academic institutions in China (Supplementary Table 1), The study workflow was shown in Fig. 1. The table 1 showed the clinicopathological characteristics of patients in the training cohort (n=799), the prospective-retrospective validation cohort (n=105), and the external validation cohort (n=180). Adjuvant chemotherapy was administered to 709 (89%) of 799 patients in the training cohort, 57 (54%) of 105 patients in the prospective-retrospective validation cohort, and 156 (87%) of 180 patients in the external validation cohort. 105 patients underwent neoadjuvant chemotherapy from the prospective-retrospective validation cohort. Median follow-up was 22.8 months (IQR 15.5–35.4) for patients in the training cohort, 24.2 months (IQR 14.3–34.9) for those in the prospective-retrospective validation cohort, and 23.0 months (IQR 9.7–48.8) for those in the external validation cohort. The detailed information regarding the patient recruitment was described in Supplementary Fig. 1.
In the univariate analysis, which was presented in Supplementary Table 2, six differentially expressed clinical characteristics were found to be associated with RFS in the training cohort, including number of tumor (P < .001), pathological N stage (P < .001), histological grade (P < .001), pathological tumor–node–metastasis (pTNM) stage (P < .001), Ki-67 status (P < .001), and Progesterone receptor status (P = .009).
Intratumoral and peritumoral signatures for predicting recurrence risk
The key radiomic features were selected by the Random forest algorithm to construct T1+C, T2WI, or DWI-ADC sequence signature by Cox regression, the detailed results were summarized in Supplementary Tables 3.
The intratumoral radiomic signature incorporated T1+C, T2WI, and DWI-ADC single sequence signature was conducted, which could assign patients into high- and low-risk groups. Patients with low-risk had better RFS in the training cohort (HR 0.06, 95% CI 0.02-0.15; P < .001), the prospective-retrospective validation cohort (P = .039), and the external validation cohort (P < .001) (Supplementary Fig. 2a-c). In addition, the efficacy of the intratumoral radiomic signature showed AUCs of 0.86, 0.88, and 0.92 for 1-, 2-, 3-year RFS prediction in the training cohort, 0.86, 0.88, and 0.86 in the prospective-retrospective validation cohort, and 0.92, 0.94, and 0.91 in the external validation cohort, respectively (Supplementary Fig. 2d-f).
Simultaneously, the peritumoral radiomic signature was constructed and also presented the ability of discriminating high- from low-risk patients in the training cohort (HR 0.04, 95% CI 0.02-0.10; P < .001), the prospective-retrospective validation cohort (P < .001), and the external validation cohort (P = .043) (Supplementary Fig. 3a-c). The peritumoral radiomic signature predicted AUCs of the 1-, 2-, and 3-year RFS of 0.96, 0.92, and 0.96 for the training cohort, 0.83, 0.86, and 0.86 for the prospective-retrospective validation cohort, and 0.87, 0.87 and 0.85 for the external validation cohort, respectively (Supplementary Fig. 3d-f).
Combined intratumoral and peritumoral signatures for predicting recurrence risk
A radiomic signature combined both intratumoral radiomic signature and peritumoral radiomic signature was developed. The intratumoral-peritumoral radiomic signature categorized patients into high- and low-risk groups, which were significantly different in terms of RFS in the training cohort (HR 0.03, 95% CI 0.01–0.07; P < .001), the prospective-retrospective validation cohort (P = .039), and the external validation cohort (P < .001) (Fig. 2a-c). Moreover, the intratumoral-peritumoral radiomic signature showed improved prediction in AUCs of the 1-, 2-, and 3-year RFS of 0.97, 0.95, and 0.98 in the training cohort, 0.86, 0.89, and 0.87 in the prospective-retrospective validation cohort, and 0.93, 0.94, and 0.91 in the external validation cohort, respectively, which presented a better predictive value than utilizing intratumoral or peritumoral radiomic signature alone (Fig. 2d-f).
In addition, the intratumoral-peritumoral radiomic signature was employed to classify high- and low-risk recurrence patients with the consideration of molecular subtype. Encouragingly, the radiomic signature could identify high- from low-risk patients in the subgroups of Luminal A (P < .001), Luminal B (P < .001), human epidermal growth factor receptor 2 (Her-2) positive (P = .007), and triple-negative breast cancer (TNBC) (P < .001) patients (Supplementary Fig. 4).
Radiomic-clinical signature for predicting recurrence risk
To develop a more precisely and clinically applicable method that could predict an individual’s recurrence, we took the clinicopathologic characteristics that associated with RFS in the univariate analysis into consideration. Multivariable analysis indicated that intratumoral-peritumoral radiomic signature, number of tumors, histological grade, pTNM stage, and Ki-67 status were independent factors of RFS (Supplementary Table 4), and these factors were used to construct the radiomic-clinical signature.
According to the radiomic-clinical signature, an optimal cutoff value (281) was generated to classify patients into high- and low-risk groups in the training cohort. A radiomic-clinical signature-print was showed to illustrate the association of these factors with the recurrence risk. The intratumoral-peritumoral radiomic signature presented the largest proportion in both high-risk (83%) and low-risk (45%) recurrence groups, followed by the histological grade (high-risk, 68%; low-risk, 44%) (Fig. 3).
This radiomic-clinical signature assigned 47 (10.4%) of 452 patients to the high-risk group, and there were significant differences in RFS between high-risk and low-risk groups (HR 0.03, 95% CI 0.01-0.08, P< .001). In the prospective-retrospective validation cohort, 40 (59.7%) of 67 patients were separated into high-risk group, which had shorter RFS (P= .030). In the external validation cohort, 34 (26.6%) of 128 patients with high-risk had shorter RFS (P < .001) (Fig. 4a-c). Besides, the radiomic-clinical signature showed better performance of RFS prediction, which achieved the higher 1-, 2-, 3-year AUCs (0.97, 0.96, and 0.98) in the training cohort, the prospective-retrospective validation cohort (AUCs of 0.88, 0.93 and 0.93), and the external validation cohort (AUCs of 0.94, 0.96 and 0.94), respectively (Fig. 4d-f).
In addition, the radiomic-clinical signature demonstrated the capability of precisely predicting recurrence risk and could be used for identifying high- and low-risk patients among different molecular subtype (P < .001 for Luminal A, P < .001 for Luminal B, P = .007 for Her2-positive, P < .001 for TNBC) (Fig. 5a-d). For Luminal subtype patients in the high-risk group who received the neoadjuvant chemotherapy showed significantly prolonged RFS (P = .048) compared with patients who received the adjuvant chemotherapy, whereas there was no added benefit of the neoadjuvant chemotherapy for patients in the low-risk group (Supplementary Fig. 5). Moreover, among Luminal subtype (T1N0M0 stage, HR-positive and Her2-negative status) patients, the radiomic-clinical signature could recognize high and low-risk patients ( P < .001; Fig. 5e), including the subgroups analysis of patients who received adjuvant chemotherapy (P < .001; Fig. 5f).
Radiomic features associated with tumor immune microenvironment and genomics
The key radiomic features from intratumoral T1+C and T2WI sequences of The Cancer Genome Altas (TCGA) and The Cancer Imaging Archive (TCIA) were found to be correlated linearly with the immune cells (Fig. 6a). The activeted natural killer cells were observed to have a positive correlation with the most radiomic features. The M0 macrophages, T cells regulatory Tregs and T cells follicular helper also presented a strong correlation. In addition, we had previously identified 29 lncRNAs which were associated with survival and immune response11. In this study, most of the lncRNAs were indicated to be remarkably correlated with the radiomic features (Fig. 6a), including the NKILA, which had been proved to play an important role in immune microenvironment in a previous study12. These results illustrated that the radiomic features could provide important information about tumor immune microenvironment.
Different classes of the radiomic features were identified using the unsupervised consensus clustering analysis in patients from TCGA and TCIA. A total of 536 differentially expressed lncRNAs and 835 differentially expressed genes were identified to be associated with radiomic features. Then the unsupervised consensus clustering analysis was performed with these lncRNAs and genes in 1,082 breast cancer patients. Two main radiomic-based lncRNA subtypes were identified to be associated with significant difference in overall survival (HR 0.71, 95% CI 0.51–0.97; P = .031) (Fig. 6b). Next, the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis were conducted to evaluate the enrichment of the radiomic-based genes. The GO enrichment analysis indicated that the radiomic-based genes were enriched in various physiological metabolic processes, such as affection of oxidoreductase activity, lipid metabolism and potassium channel complex, details were illustrated in Fig. 6c. The KEGG pathway enrichment analysis found these genes were involved in the vitamin digestion and absorption and peroxisome proliferator-activated receptor signaling pathway.