Prognostic impact of EGFR mutations in T1–4N0M0 lung adenocarcinoma: analyses focus on imaging and pathological features Authors: Jing-Yu Chen, Ying Zhu, Bao-Cong Liu, Hui-Yun Ma, Lu-Jie Li, Mei-Cheng Chen, Shu-Chang Zhou, Xiang-Min Li, Jian-Ting Long, Qiong Li Published: 2024 ABSTRACT BackgroundWith the development of tyrosine kinase inhibitor (TKI) treatment, the prognosis of advanced lung adenocarcinoma (LUAD) patients with epidermal growth factor receptor (EGFR) mutations has been continuously improving. This study aims to propose the utilization of pathological characteristics and imaging features to evaluate the impact of EGFR gene mutations on the prognosis of T1–4N0M0 LUAD.MethodsAmong the cases diagnosed with LUAD between April 2015 and April 2016, 438 patients with T1–4N0M0 LUAD were included, and the clinical characteristics were collected. EGFR mutations were analyzed in these patients who underwent lobectomy with different radiological and pathological types for the relation to patient prognosis.ResultsPatients with EGFR mutation had longer recurrence-free survival (RFS) in part-solid nodules cohort (P=0.03), which was in contrast to purely solid nodules (P=0.06). Positive EGFR mutations significantly prolonged RFS in nodules consolidation-to-tumor ratio (CTR) values of 0–0.5. In the International Association for the Study of Lung Cancer (IASLC) grade I patients with EGFR mutations, there was a trend towards longer RFS but with no effect on overall survival (OS) (P=0.08; P=0.71); in IASLC grade II patients with EGFR mutations, there was a tendency of longer OS (P=0.06); in IASLC grade III patients with EGFR mutations, both RFS and OS were significantly shorter (P=0.02; P=0.005). EGFR mutation state was not an independent risk factors for both RFS and OS.ConclusionsEGFR mutations are associated with a favorable prognosis in nodules with lower IASLC grading or more ground glass opacity (GGO) components. The results were reversed in patients with higher IASLC grading or no GGO component. Highlight box Key findings • Epidermal growth factor receptor (EGFR) mutation indicates a better prognosis in nodules with low International Association for the Study of Lung Cancer (IASLC) grades or ground glass opacity (GGO) components. • Higher IASLC grade nodules or absence of GGO components reverse the favorable EGFR prognosis. What is known and what is new? • The prognosis of invasive lung adenocarcinoma (LUAD) is influenced by pathological, imaging, and molecular characteristics. What is the implication, and what should change now? • Integrating radiologic features and molecular biomarkers into the staging system for invasive LUAD could improve diagnostic accuracy and support precision medicine. Introduction Epidermal growth factor receptor (EGFR) mutations are prevalent in lung adenocarcinoma (LUAD), with a noninterventional real-world study (ICAN) demonstrating a high rate of 55.1% in Chinese patients with resectable LUAD. These mutations have been identified as a significant independent prognostic factor for overall survival (OS) (1). The introduction of tyrosine kinase inhibitor (TKI) therapy has led to notable advancements in the treatment outcomes of advanced LUAD patients harboring EGFR mutations (2,3). The impact of EGFR mutations on the prognosis of surgically resected patients has been a subject of debate. While several studies, including a large retrospective study, have reported a significant association between EGFR mutations and improved prognosis (1,4), conflicting findings have been observed in other studies (5,6). These discrepancies can be attributed to variations in tumor histological subtypes, biological behaviors, and radiological presentations, which introduce confounding factors that are difficult to control. In 2020, the International Association for the Study of Lung Cancer (IASLC) Pathology Committee proposed a new grading system for LUAD, classifying invasive pulmonary adenocarcinoma into three grades based on the proportion of acinar components, solid components, papillary structures, and the presence of necrosis (7). In patients with the IASLC grades 2 and 3, carriage of EGFR and Kirsten rat sarcoma viral oncogene homologue (KRAS) mutations has been shown to be associated with a worse prognosis (8,9), further research is still needed to better understand the impact of EGFR mutations in this context and to elucidate the interplay between EGFR mutations, ground glass opacity (GGO) components, and IASLC pathological grades in determining lung cancer prognosis. In this study, our aim was to conduct a comprehensive investigation into the prognostic value of EGFR mutations in T1–4N0M0 LUAD with diverse pathological grades and imaging features. We aspire to unlock new frontiers in understanding the intricate interactions between EGFR mutations and the different features of LUAD. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-724/rc). Methods Cohorts and clinical characteristics Patients with invasive LUAD (T1–4N0M0) who underwent lobectomy and received EGFR mutation detection were collected from Sun Yat-sen University Cancer Center and the First Affiliated Hospital of SYSU between April 2015 and April 2016. This study was conducted in accordance with the Helsinki Declaration (revised in 2013) and the Institutional Review Board of Sun Yat-sen University Cancer Center approved this study (B2022-293-01). Informed consent was waived due to the retrospective nature of this study. All cases included in the study were treatment-naïve and underwent enhanced chest computed tomography (CT) examinations upon admission. Metastasis was excluded based on contemporaneous CT/magnetic resonance imaging (MRI)/positron emission tomography (PET)/CT scans or histopathological examination, as shown in Figure 1. Postoperative pathological diagnosis was conducted according to the criteria of the IASLC/American Thoracic Society (10). Patient selection flowchart. LUAD, lung adenocarcinoma; CT, computed tomography; EGFR, epidermal growth factor receptor. EGFR mutation analysis EGFR mutations in exons 18, 19, 20, and 21 were analyzed through a real-time polymerase chain reaction (RT-PCR)-based amplification refractory mutation system using the Super-ARMS® EGFR Mutation Detection Kit. Based on the outcome of the EGFR test results, the patients were divided into two groups: the EGFR mutation group and the EGFR wild-type group. The EGFR mutation sites detected in this study are displayed in Table S1. Histological evaluation and definition of the IASLC grading system All tumor slides from eligible patients were reviewed and classified according to the proposed IASLC grading system by two thoracic pathologists blinded to the patients’ clinical outcomes. As referenced in the IASLC grading system (7). Pleural invasion was defined as invasion of the tumor beyond the elastic layer or to the pleural surface but not to the parietal pleura. Vascular infiltration and nerve invasion was defined by tumor cells in the lumen of vessels or adjacent to the nerve bundles, respectively. The IASLC grading system includes the following three grades: grade 1 (G1; i.e., lepidic predominant tumor with <20% high-grade patterns), grade 2 (G2; i.e., acinar or papillary predominant tumor with <20% high-grade patterns), and grade 3 (G3; i.e., any tumor with ≥20% high-grade patterns). Scan protocol All preoperative chest CT images were acquired using multidetector CT scanners, including Aquilion 64 (Canon Medical Systems, Otawara, Japan), CT750 (GE Healthcare, Milwaukee, WI, USA), and Siemens SOMATOM Force CT scanner (Erlangen, Germany), with patients in an inspiratory state. The scan parameters consisted of a tube voltage of 120 kVp, a maximum of 200 mA with automatic tube current modulation, a field of view (FOV) of 350 mm, and an image matrix of 512×512. Axial thin-section CT images of the entire lung were reconstructed with a section thickness of 1.0 mm using a high-resolution algorithm without any intervals. The clinical data from Sun Yat-sen University Cancer Center used in this study were deposited on the Research Data Deposit public platform (https://www.researchdata.org.cn/) under the approval number RDDA2019001083. Semantic CT characteristics Semantic CT imaging characteristics for evaluation included lesion location, size, density, shape, lobulation, speculation, consolidation-to-tumor ratio (CTR), air bronchogram, bubble-like lucency, and pleural indentation sign. The definitions of specific CT features are outlined in Appendix 1. Enhanced chest CT images were acquired within 1 week prior to treatment and the CT imaging features were evaluated by three experienced chest radiologists (XY Yang, DD Chang, and X Wu, with experience of 21, 13, and 25 years respectively) through picture archiving and communication system (PACS) reading workstation blinded to gene mutation status to control potential bias. The order of all patients was disrupted during analysis. Consensus was reached when the radiologists disagreed. It is important to verify the prognostic impact of EGFR in patients with different solid component proportions. The imaging physicians divided the patients into 0< CTR ≤0.5, 0.5< CTR <1, and CTR =1 groups based on their baseline CT images to explore the prognostic impact of EGFR mutation status in different patients. Follow-up protocol Routine follow-up of patients commenced after the surgical procedure. Chest CT scans, abdominal and cervical/supraclavicular ultrasonography, and MRI or CT scans of the brain were conducted every 4 months for the initial 3 years post-surgery, every 6 months for the subsequent 2 years, and annually thereafter. Bone scans were performed once a year. The sites of initial recurrence were categorized as “thorax”, “abdomen”, “neck”, “brain”, and “bone” based on the diagnostic tests conducted during follow-up. Recurrence-free survival (RFS) was defined as the duration from the date of surgery to the date of first recurrence or last follow-up. Patients who died from causes unrelated to invasive LUAD were considered censored in the RFS analysis. OS was defined as the time from the date of surgery to the date of death or last follow-up. Statistical analysis Baseline characteristics were reported as number for categorical variables. Estimation of survival curves of RFS and OS were generated by the Kaplan-Meier method; the log-rank test was used to compare survival curves. Cox regression modeling was used as the method for identifying risk factors for LUAD. All P values were two-sided with a significance level of 0.05. All statistical analysis were performed by using SPSS software (version 21.0; IBM Corporation, Armonk, NY, USA) and PRISM software (version 9.3.1; GraphPad Software, La Jolla, CA, USA). Results Basic clinical and semantic CT imaging characteristics A total of 438 patients who underwent through resection for T1–4N0M0 LUAD were included in our cohort. The clinical characteristics of all patients are presented in Table 1. Among the patients, 216 (49.3%) were male and 222 (50.7%) were female, while 190 (43.4%) patients were aged over 60 years. Regarding tumor size, 348 (79.5%) patients had nodules with a maximum diameter of less than 3 cm, classified as stage T1, while 90 (20.5%) patients had nodules with a maximum diameter greater than 3 cm. Pathological examination revealed pleural invasion, nerve invasion, and vascular infiltration in 108, 5, and 25 patients, respectively. Based on the IASLC grading system, G1 tumors were observed in 24.4% (n=107) of the cases, G2 in 58.0% (n=254), and G3 in 17.6% (n=77). On preoperative CT images, the nodule types were categorized as pure ground-glass nodules (pGGNs) in 22 patients, mixed ground-glass nodules (mGGNs) in 137 patients, and solid tumors in 279 patients. Additionally, spiculation sign was observed in 186 patients, while different degrees of pleural indentation were noted in 346 patients. Air bronchogram sign was detected in CT images of 120 patients, and abnormal bronchial sign was observed in 55 patients. EGFR gene mutation testing was performed for all patients, with 287 patients showing EGFR mutation, including 116 patients with exon 19 deletion and 147 patients with exon 21 mutation. The associations of EGFR mutation status with each pathologic and radiology characteristics is demonstrated in Table S2. Table 1Patient characteristicsVariablesCases (n=438)Clinical variables Sex Male216 Female222 Age (years) <60248 ≥60190 Smoking status Never362 Ever/current76Pathological variables IASLC grade Grade I107 Grade II254 Grade III77 Pleural invasion Yes108 No330 Nerve invasion Yes5 No433 Vascular infiltration Yes25 No413 EGFR mutation No mutation151 19 exon deletion116 21 exon mutation147 Others24Radiological variables T descriptor T1332 T295 T310 T41 Nodule type pGGN22 mGGN137 Solid279 CTR 0< CTR ≤0.561 0.5< CTR <198 CTR =1279 Lobulation Yes424 No14 Spiculation Yes186 No252 Pleural indentation None92 Pleural contact40 Pleural thickening11 Pleural retraction with 1 stripe85 ≥2 stripes or pleural tags210 Air bubble sign or abnormal bronchial sign None263 Air bubble sign120 Abnormal bronchial sign55IASLC, International Association for the Study of Lung Cancer; EGFR, epidermal growth factor receptor; pGGN, pure ground-glass nodule; mGGN, mixed ground-glass nodule; CTR, consolidation-to-tumor ratio. Patient characteristics IASLC, International Association for the Study of Lung Cancer; EGFR, epidermal growth factor receptor; pGGN, pure ground-glass nodule; mGGN, mixed ground-glass nodule; CTR, consolidation-to-tumor ratio. Correlation between EGFR and prognosis For the entire cohort, the median follow-up duration was 63.0 months [95% confidence interval (CI): 60.9–65.1], and the 5-year RFS and OS rates were 86.02% and 92.57%, respectively. We examined how EGFR mutations affect RFS and OS in patients with various nodule types (Figure 2). No significant differences were observed in RFS and OS between patients with EGFR mutations and those with EGFR wild type (Figure 2A,2B). Patients with EGFR mutation had longer RFS in part-solid nodules cohort (P=0.03), which was in contrast to the results in patients with purely solid nodules (Figure 2C,2E; P=0.06). The presence or absence of an EGFR mutation was not observed as a significant difference in OS between the two cohorts of patients (Figure 2D,2F). Kaplan-Meier curves of EGFR mutation vs. wild type for all patients’ RFS (A) and OS (B); patients with part-solid nodules’ RFS (C) and OS (D); patients with pure-solid nodules’ RFS (E) and OS (F). EGFR, epidermal growth factor receptor; RFS, recurrence-free survival; OS, overall survival. Patients were classified into 0< CTR ≤0.5, 0.5< CTR <1, and CTR =1 groups according to the percentage of ground glass component in the nodules to be analyzed. Among the patients in the 0< CTR ≤0.5 group, patients with EGFR mutations had significantly longer RFS (Figure 3A; P=0.01), while no significant difference was observed in OS (Figure 3B). Neither RFS nor OS differences were significant in patients in the 0.5< CTR <1 group (Figure 3C,3D). Kaplan-Meier curves of EGFR mutation vs. wild type of RFS (A), OS (B) in patients with 0< CTR ≤0.5 and RFS (C) and OS (D) in patients with 0.5< CTR <1. CTR, consolidation-to-tumor ratio; EGFR, epidermal growth factor receptor; RFS, recurrence-free survival; OS, overall survival. EGFR has different effects in different IASLC classifications We compared the prognostic impact of EGFR mutation frequency in patients with different IASLC grades, as shown in Figure 4. It was evident that EGFR mutations were significantly decreased in patients with IASLC grade III compared to patients with grades I and II. In addition, we could see that in patients with grade I, RFS was significantly prolonged in EGFR mutation positive patients (Figure 4A; P=0.008), in patients with grade II, no relevant trend could be observed, while in patients with grade III, both RFS and OS prognosis were significantly shorter in EGFR mutation positive patients (Figure 4E,4F; P=0.02; P=0.04), suggesting a different role of EGFR mutations in patients with different pathological grades. In addition, the frequency of EGFR mutations was notably higher in patients with lower IASLC classification (Figure S1). Kaplan-Meier curves of EGFR mutation vs. wild type of RFS (A), OS (B) in patients with IASLC grade I, RFS (C) and OS (D) in patients with IASLC grade II and RFS (E) and OS (F) in patients with IASLC grade III. IASLC, International Association for the Study of Lung Cancer; EGFR, epidermal growth factor receptor; RFS, recurrence-free survival; OS, overall survival. Given the substantial difference in IASLC grade between solid nodules and GGN-containing nodules, we conducted a separate analysis to investigate the influence of EGFR mutations on solid nodules across different IASLC grades (Figure S2), and it was found that there was a slight variation in the relationship between EGFR mutations and IASLC grading in solid nodules as well. We subsequently performed a separate analysis to assess the impact of EGFR mutations on prognosis across different IASLC grade nodules, as shown in Figure 5. It could be simply found that in patients with IASLC grade I and grade III, the prognostic impact of EGFR mutations in patients with solid nodules follows the same trend as in the total population. In IASLC grade I patients with EGFR mutations there was a trend towards longer RFS and no effect on OS (Figure 5A,5B; P=0.08; P=0.71); in IASLC grade III patients with EGFR mutations both RFS and OS were significantly shorter (Figure 5E,5F; P=0.02; P=0.005). In contrast, in patients with IASLC grade II, the differences in RFS and OS between EGFR mutation-positive and wild-type patients were not significant, but a trend towards shorter RFS (Figure 5C; P=0.08) and longer OS (Figure 5D; P=0.09) could be observed, which was different from the trend in the total population. Kaplan-Meier curves of EGFR mutation vs. wild type of RFS (A), OS (B) in patients with solid nodules in IASLC grade I, RFS (C) and OS (D) in patients with solid nodules in IASLC grade II and RFS (E) and OS (F) in patients with solid nodules in IASLC grade III. IASLC, International Association for the Study of Lung Cancer; EGFR, epidermal growth factor receptor; RFS, recurrence-free survival; OS, overall survival. Univariate and multifactorial analyses of RFS and OS Univariate analysis and Cox regression analysis were performed on RFS and OS of patients, respectively (Table 2), and found that high IASLC grade, pleural invasion, nodules with solid component greater than 3 cm, solid nodule and nodule with spiculation had a negative impact on RFS. Pleural invasion and solid nodules being independent risk factors for RFS. In contrast, age over 60 years, solid component greater than 3 cm, solid nodules, and spiculation were negative factors for OS. Age greater than 60 years and solid components being an independent risk factor for OS. Table 2Univariate analysis and Cox regression analysis of factors associated with RFS and OS for patientsVariablesUnivariate P valueMultivariate analysisHR (95% CI)PRFS Gender (female vs. male)0.820.67 Age (≥60 vs. <60 years)0.410.23 Smoking history (yes vs. no)0.95 IASLC grade (G1 vs. G2 vs. G3)0.059 Pleural invasion (yes vs. no)0.0012.049 (1.245, 3.367)0.005 Nerve invasion (yes vs. no)0.79 Vascular infiltration (yes vs. no)0.93 Solid components (<3 vs. ≥3 cm)<0.0010.43 Nodule type (pGGN and mGGN vs. solid)0. 0013.876 (1.901, 7.874)<0.001 Lobulation (yes vs. no)0.46 Spiculation (yes vs. no)0.0010.36 EGFR mutation (yes vs. no)0.72OS Gender (female vs. male)0.20 Age (≥60 vs. <60 years)0.031.972 (1.026, 3.788)0.042 Smoking history (yes vs. no)0.21 IASLC grade (G1 vs. G2 vs. G3)0.50 Pleural invasion (yes vs. no)0.49 Nerve invasion (yes vs. no)0.55 Vascular infiltration (yes vs. no)0.58 Solid components (<3 vs. ≥3 cm)<0.0010.40 Nodule type (mGGN and pGGN vs. solid)0.0027.042 (2.155, 22.73)0.001 Lobulation (yes vs. no)0.65 Spiculation (yes vs. no)0.0070.45 EGFR mutation (yes vs. no)0.49RFS, recurrence-free survival; OS, overall survival; HR, hazard ratio; CI, confidence interval; IASLC, International Association for the Study of Lung Cancer; G1, grade 1; G2, grade 2; G3, grade 3; mGGN, mixed ground-glass nodule; pGGN, pure ground-glass nodule; EGFR, epidermal growth factor receptor. Univariate analysis and Cox regression analysis of factors associated with RFS and OS for patients RFS, recurrence-free survival; OS, overall survival; HR, hazard ratio; CI, confidence interval; IASLC, International Association for the Study of Lung Cancer; G1, grade 1; G2, grade 2; G3, grade 3; mGGN, mixed ground-glass nodule; pGGN, pure ground-glass nodule; EGFR, epidermal growth factor receptor. Discussion To further investigate and validate the impact of EGFR mutations on the prognosis of patients undergoing surgical resection for T1–4N0M0 LAUD, our study investigated prognostic variations in patients with resectable primary LUAD characterized by different nodule types, IASLC grading, and EGFR mutation status. Consistent with previous findings, EGFR mutations may not influence RFS and OS in the overall patient cohort. However, we observed intriguing results within specific subgroups. Particularly, patients with EGFR mutations in part solid nodules showed a longer RFS compared to those with wild-type EGFR. Conversely, EGFR mutations in solid nodules were associated with a favorable prognosis in low IASLC grades but a poorer outcome in high IASLC grades. Further investigation is warranted to elucidate the mechanisms underlying these relationships and to tailor treatment approaches accordingly. In 2020, the IASLC introduced a novel pathological grading system that categorizes stage I–III invasive LUAD into three grades based on the histological characteristics of cancer cells and the prevalence of high-grade cell types. This grading system correlates lower cancer cell differentiation and a higher proportion of high-grade pattern cells with decreased OS rates in patients (7). In advanced LUAD, worse pathological grade was associated with worse prognosis (11). Another study found that lower pathological grade may be more likely to have EGFR, KRAS mutations and anaplastic lymphoma kinase (ALK) fusions, and that patients with EGFR mutations in advanced stage had worse outcomes from chemotherapy in the low-grade group compared to those with high-grade LUAD (12). In a recent study, EGFR mutations were shown to be associated with a shorter prognosis in patients undergoing IASLC grade III surgery for LAUD cancer (8). However, limited by the insufficient number of studies, our understanding of the association between EGFR mutations and the IASLC grading system in patients undergoing surgery for T1–4N0M0 LUAD is still incomplete. This study investigated the prognostic implications of EGFR mutations in patients with varying IASLC grades. Our findings indicate that patients with EGFR mutations had improved RFS in grade I patients, but both RFS and OS were notably shorter in EGFR mutation-positive patients with grade III disease. These results align with previous research in this area. However, there is a lack of studies investigating the correlation between IASLC grade, the presence of GGO components, and the prognostic implications of GGO components in tumors of different IASLC grades. We found that nodules containing GGO had a relatively lower IASLC grade and that patients with EGFR mutations may have a better prognosis in tumors containing GGO, whereas in solid tumors, the relationship between EGFR mutations and IASLC grading showed some inconsistencies. Specifically, among grade II patients, the disparities in RFS and OS between EGFR mutation-positive and wild-type patients were not statistically significant. Nevertheless, a tendency towards shorter RFS and longer OS was observed, diverging from the patterns observed in the general population. We therefore infer that the biological behavior of GGO-containing tumors may be different from that of solid tumors, and different considerations need to be made when treating them. However, in the 8th edition of staging, the exclusion of the GGO component when measuring tumor size is proposed (10,13). A recent study has shown that patients with EGFR mutations may have a lower risk of recurrence for nodules containing a GGO component, while no similar phenomenon was observed in pure solid tumors (14). Previous research indicates that GGO-containing tumors have a higher incidence of EGFR mutations compared to solid tumors. Additionally, a higher percentage of GGO is associated with a higher likelihood of EGFR mutation, but a lower frequency of EGFR amplification. This suggests that EGFR may influence the transformation of GGO to solid components, but the correlation between EGFR mutation in the percentage of GGO and prognosis remains unclear (15-18). The CTR is an important indicator that reflects the proportion of GGO component within nodules, and it can provide clinical guidance for surgical considerations (19-21). A previous study has shown that tumors with EGFR mutations tend to have lower CTR values. However, the prognostic impact of EGFR mutations in nodules with different CTR values has not been extensively investigated (22). In our study, we observed a significant prolongation of RFS in patients with EGFR mutations in nodules with 0< CTR ≤0.5. However, in nodules with 0.5< CTR <1, EGFR mutations may not affect RFS and OS. These findings suggest that a higher percentage of ground glass components in nodules may have a more favorable impact on patients with EGFR mutations. Despite the wide use of the current 8th edition of staging system, it exhibits notable limitations. In the context of diagnosing and treating invasive LUAD, factors such as the proportion of GGO components in nodules, the status of EGFR mutations, and the IASLC grading system hold significant prognostic implications. GGO components are generally correlated with reduced tumor invasiveness and improved prognosis, particularly among patients with EGFR mutations. However, the prognostic significance of varying proportions of GGO components remains uncertain. The IASLC grading system enhances prognostic precision through histological assessment; however, it remains deficient in accounting for the biological behavior of tumors with GGO components. Consequently, integrating radiologic features and molecular biomarkers into the staging system for invasive LUAD could potentially improve diagnostic and prognostic accuracy. Such an approach may also support precision medicine in the management of invasion LUAD. There are some limitations in this study. Firstly, the failure to account for the presence of spread through air spaces (STAS) during data collection, a crucial element in the new IASLC classification, may have impacted the accuracy of the findings (23,24). Secondly, while the study encompassed 438 patients from various institutions, the necessity for validation in a larger, meticulously designed prospective study is imperative due to the inherent constraints of retrospective studies. Conclusions In conclusion, our research finds that EGFR mutations correlate with prognosis in invasive LUAD, showing better outcomes in low IASLC grade or GGO nodules and worse outcomes in high IASLC grade or solid nodules. We believe that a more detailed staging system incorporating EGFR mutations, pathological features, and radiology features may better guide clinical diagnosis and treatment. Supplementary The article’s supplementary files as Available at https://jtd.amegroups.com/article/view/10.21037/jtd-24-724/prf