Abstract
Background/Aim: For patients with T1a muscularis mucosae (MM) esophageal squamous cell carcinoma (ESCC) with lymphovascular invasion (LVI) or T1b submucosal (SM) ESCC, endoscopic resection is non-curative, and adjuvant treatment entailing esophagectomy or definitive chemoradiotherapy is necessary. This is because about 30% of these cases have lymph node (LN) metastasis. The purpose of this study was to test the utility of a CRP genetic polymorphism test kit for determining the risk of LN metastasis with the aim of eliminating additional invasive adjuvant therapy. Patients and Methods: This is a retrospective, multi-institutional, observational study. The CRP 1846C>T genetic polymorphisms were identified using a fully automated genotyping system. The primary end points were an 85% negative predictive value (NPV) for diagnosis of LN metastasis in pT1a (MM) and 80% NPV in pT1b (SM1) patients. Results: A total of 742 ESCC (105 pMM, 166 pSM1 and 471 pSM2-3) patients who had received esophagectomy with 2- or 3-field LN dissection at 65 institutions were enrolled. According to this test, patients with the C/C and C/T genotypes were considered to be low risk. The NPVs using this test were 82.8% in pMM and 71.7% in pSM1 patients. Conclusion: CRP 1846C>T genetic polymorphism is not a useful diagnostic indicator for determining the risk of LN metastasis; however, the possibility that CRP gene polymorphisms are involved in the mechanism of lymph node metastasis in solid tumors still remains.
More than 570,000 patients are diagnosed with esophageal cancer each year worldwide, and the dominant histological subtype is esophageal squamous cell carcinoma (ESCC) (1). The prognoses for patients with advanced ESCC are poor, despite treatment with intensive multimodal therapies (2). However, cStage I (cT1N0M0) ESCC can be cured with esophagectomy with lymph node (LN) dissection, after which the 5-year survival rate is more than 80% (2, 3). It is noteworthy, that more than 30% of esophagectomies in Japan are for superficial thoracic ESCC (2). Unfortunately, some of these patients succumb to operation-related death or suffer from serious complications, even at cStage I (2, 3). Moreover, after esophagectomy, patients often suffer persistent problems with functional symptoms (e.g., reduced food intake, frequent regurgitation and dyspnea), which significantly reduce their quality of life (QOL) (4-6). Because of these, definitive chemoradiotherapy (CRT) was developed as a primary organ-preserving alternative to esophagectomy for patients with cStage I (cT1bN0M0) ESCC (3). Among these patients, the 5-year overall survival rate is 85%, but grade 3-4 acute adverse events, including neutropenia, leukocytopenia, hyponatremia, and esophagitis, among others, occur in up to 11% of patients (3). Furthermore, late adverse events, including cardiac ischemia, dyspnea, and pleural effusion, occur in 2-3% of patients, reducing their QOL to the same degree as esophagectomy (3). There is thus a great need for a reliable, curative, and less invasive treatment for cStage I (cT1bN0M0) ESCC.
Recent endoscopic imaging technologies have enabled early detection of ESCC, and use of endoscopic resection (ER), mainly endoscopic submucosal dissection (ESD), is rapidly becoming more widespread (7-9). This is due in large part to improvements in endoscopic technology as well as to the increased experience of endoscopists. At present, ER is absolutely indicated for patients with T1a mucosal ESCC without lymphovascular invasion (LVI) (10). However, among patients with T1a muscularis mucosae (MM) ESCC with LVI or T1b submucosal (SM) ESCC, ER is recognized to be non-curative, even when the vertical and horizontal margins are negative, and the preoperative diagnosis is negative for LN involvement (10-14). The potential presence of undetectable LN metastasis prompts us to be aggressive, administering esophagectomy with extended lymphadenectomy or definitive CRT. To extend the indication for ER, especially ESD, for T1a (MM) or T1b submucosal ESCC, which would be highly desirable for oncological safety, the key is to accurately diagnose the risk of LN involvement. Conventional examination for LN metastasis entails contrast-enhanced computed tomography (CE-CT), endoscopic ultrasonography (EUS), ultrasonography from the body surface (US), and/or 18F-2-fluoro-2-deoxy-D-glucose-positron emission tomography-CT (PET-CT) (15-17). Among these, CE CT has a sensitivity of 40-60% and an accuracy of 50-70% for diagnosis of LN metastasis (15-17). Negative predictive values (NPVs) are reportedly 58%, 60%, and 63% for CE CT, EUS, and PET-CT, respectively (15-17).
We previously reported that LN involvement in thoracic ESCC may have a genetic component − i.e., CRP 1846C>T genetic polymorphism appears to account for more than 80% of the NPV when diagnosing the risk of LN metastasis in T1b submucosal ESCC (18, 19). The present study is a multicenter retrospective observational study for development of a new test kit using CRP 1846C>T genetic polymorphism to assess the risk of LN metastasis in pT1a MM with LVI and pT1b ESCC. With implementation of this test for cN0 patients, in the event of pathological findings showing pT1a (MM) with LVI and pT1b (SM) ESCC, one would be able to assess the risk of LN metastasis. The results of the test, with which a negative result indicates a low risk of metastasis, will help the decision whether or not to complete the treatment using only ER or proceed to more invasive treatments. By providing information about the risk of LN metastasis and the appropriate treatment, this test would help make it possible to carry out more optimized individual medicine (precision medicine) for patients.
Our findings from this study aimed to determine whether the ESD indication can be safely extended to T1a (MM) with LVI and T1b (SM) ESCC showed that our new test kit for the CRP 1846C>T genetic polymorphism provides a high NPV.
Patients and Methods
Study design. This study (UMIN-Clinical Trials Registry no; UMIN000038043) is a retrospective, multi-institutional, observational study conducted in accordance with the Ethical Guidelines for Medical and Health Research Involving Human Subjects, Ethical Guidelines for Human Genome/Gene Analysis Research, and the principles of the Declaration of Helsinki. The protocol was approved by the Tohoku University Hospital (No. 11000628) as a collective ethical reviewer or by the respective institutional review board of the ethics committee at each site. All patients provided written informed consent.
Data management. Clinical data management and monitoring for the quality control of this study were conducted by the Clinical Research Data Center of Tohoku University Hospital. Quality control was ensured through the use of pre-designed monitoring procedures and a data management plan, and study data were collected using Electronic Data Capture (EDC). For cases for which consent has been obtained, researchers registered the subjects and entered data on an EDC system using their own accounts. The principal investigator requested that data managers and monitoring managers monitor the data to confirm that they were collected accurately.
Pathologic depth of invasion of submucosal cancer and abbreviations. Based on the guidelines for clinical and pathologic studies on carcinoma of the esophagus, submucosal infiltration was divided into three stages, depending on the depth of infiltration in the surgically resected specimen (10). These are abbreviated as follows: SM1 is cancer infiltrating the superficial one-third of the submucosal layer; SM2 is cancer infiltrating the middle one-third of the submucosal layer; and SM3 is cancer infiltrating the deepest one-third of the submucosal layer. Because this classification is not applied to ER samples from patients who received ESD as a primary treatment before esophagectomy, a lesion with a depth of <200 μm from the lower surface of the MM was classified as SM1, and a depth >200 μm was classified as SM2.
Inclusion and exclusion criteria. To be included in this study, patients had to be at least 20 years of age and have a pathological diagnosis of thoracic ESCC. After being diagnosed with cN0 (without LN metastasis) and cM0 ESCC based on CE-CT, EUS, US, and/or PET-CT, patients received right transthoracic or thoracoscopic esophagectomy (R0) with 2- (mediastinal and abdominal) or 3-field (bilateral neck, mediastinal and abdominal) LN dissection. After esophagectomy, patients with tumors pathologically diagnosed as pT1a (MM) with LVI or pT1b (SM1 and SM2-3) with or without LVI were enrolled. Consequently, some patients were diagnosed as pN0, whereas others were pN1. The disease was classified according to the UICC International Union Against Cancer Tumor-Node-Metastasis (TNM) Classification of Malignant Tumors (7th Edition).
Analysis of CRP genetic polymorphism. Peripheral blood samples were collected from the patients and stored at −20°C at each site until transferred to a laboratory at Akita University Graduate School of Medicine. After allowing the whole blood to thaw, we investigated CRP 1846C>T (rs1205) genetic polymorphisms using a fully automated genotyping system (M.K and S.M) (i-densy IS-5320; Arkray Inc., Kyoto, Japan). In addition, after extracting DNA from samples of pT1a (MM) and pT1b (SM1) ESCC using a QIAamp Blood kit (Qiagen, Hilden, Germany), CRP 1846C>T (rs1205) genetic polymorphism was investigated using the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method (K.F). The methods used to identify genotypes and details of the primers and PCR conditions have been published elsewhere (18). The DNA samples were also analyzed using the fully automated i-densy IS-5320 genotyping system to confirm the agreement between the two methods (M.K and S.M).
The patients were divided based on whether they were diagnosed with or without pathological LN metastasis (pN0 and pN1) and whether they carried the CRP 1846C/C+C/T or T/T genotype, after which the influence of CRP 1846C>T polymorphism on LN involvement in thoracic ESCC was investigated. Based on earlier studies, we defined 1846C/C+C/T as low risk for LN metastasis and T/T as high risk (18, 19). Even when the pathological diagnosis of LN metastasis was negative, if there was a diagnosis of LN metastasis or recurrence of esophageal cancer within 2 years after surgery, the sample was defined as positive for pathological LN metastasis, as it is highly likely that metastasis was initially overlooked.
Endpoints and assessments. The primary end points were the NPV for diagnosis of LN metastasis using the CRP1846C>T genetic polymorphism test in patients with pT1a (MM) or pT1b (SM1) thoracic ESCC. Secondary end points included NPV for diagnosis of LN metastasis in patients with pT1b (SM2-3) and the positive predictive values (PPVs), sensitivities and specificities for LN metastasis in all three groups. These analyses were performed with patients who received ESD as a primary treatment before esophagectomy.
Factors analyzed in this study consisted of 1) basic patient information from their preoperative evaluations and blood tests, including CRP (mg/dl); 2) surgical information, including fields of LN dissection, numbers of LNs dissected from the neck, chest, and abdomen, and total numbers of LNs dissected; and 3) pathological diagnosis information, including pN, grade of differentiation, lymphatic invasion (ly), venous invasion (v), and 4) postoperative recurrence.
Statistical analysis. All analyses were prespecified in the statistical analysis plan. Target sample sizes were planned as follows: the expected and threshold value of the NPV using the new kit was set at 0.95 and 0.85, respectively, for the pT1a (MM) group and at 0.90 and 0.80 for the pT1b(SM1) group. Consequently, a sample size of 105 in the pT1a (MM) group and 140 in the pT1b (SM1) group was required for a one-sided α of 0.025 and power of 80%. In the pT1b (SM2-3) group, the expected and threshold NPVs with this kit were set at 0.85 and 0.80, respectively, and the required sample size was 400.
For patient backgrounds, discrete variables are shown as frequencies and percentages, while continuous variables are shown as medians with minimum-maximum. NPVs, PPVs, sensitivities, and specificities are presented as point estimates with two-sided 95% exact confidence intervals, which are based on the Clopper Pearson method. The significance level was set to one-sided 0.025, and the p-value was calculated based on a binomial distribution under the null hypothesis (threshold value for PPV=30%, sensitivity=40%, specificity=65%). Prespecified subgroup analysis of the primary endpoint was also conducted for ER margin positive and ER margin negative subgroups. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
For this study, 866 patients from 65 institutions were enrolled between October 9, 2019, and Jan 31, 2022. Of these patients, 118 ineligible cases and 6 cases in whom CRP genetic polymorphism could not be determined were excluded. The remaining 742 ESCC patients, among whom 105 were pT1a (MM), 166 were pT1b (SM1), and 471 were pT1b (SM2-3) ESCC, were analyzed.
The backgrounds of patients segregated into the pT1a (MM), pT1b (SM1), and pT1b (SM2-3) groups are shown in Table I. For 30% of patients in the pT1a (MM) and pT1b (SM1) groups and 18% of patients in the pT1b (SM2-3) group, ESD was performed as the primary treatment before esophagectomy. The numbers of dissected LNs were 48-53. Pathological LN metastasis was detected in 22-28% of patients [the median number was 1 (1-7)]. Lymphatic invasion was detected in 34.3% of pT1b (SM1) and 43.7% of pT1b (SM2-3) patients.
Background of patients segregated into 3 groups.
The frequencies of the CRP 1846C>T (rs1205) genotypes were all consistent with what would be expected if the population was in Hardy-Weinberg equilibrium and were similar to the genotype frequencies reported in the National Cancer Institute SNP500 database, as were the respective genotype frequencies in the pT1a (MM), pT1b (SM1), and pT1b (SM2-3) groups (Table I).
When investigating CRP 1846C>T (rs1205) genetic polymorphisms, genotypes determined using the PCR-RFLP method matched completely with those determined using the fully automated i-densy IS-5320 genotyping system in all 742 cases.
The primary end points, NPVs, in the pT1a (MM) and pT1b (SM1) groups were 82.8% and 71.7%, respectively, which did not meet the hypothesis (Table II and Table III). In the pT1b (SM2-3) group, the NPV for LN metastasis was 72.1% (Table IV).
Relationship between C-reactive protein (CRP) genetic polymorphism and pathological lymph node (LN) metastasis in pT1a [muscularis mucosae (MM)] patients.
Relationship between C-reactive protein (CRP) genetic polymorphism and pathological lymph node (LN) metastasis in pT1b [submucosal 1 (SM1)] patients.
Relationship between C-reactive protein (CRP) genetic polymorphism and pathological lymph node (LN) metastasis in pT1b [submucosal 2-3 (SM2-3)] patients.
As mentioned above, about 30% of pT1a (MM) and pT1b (SM1) patients and 18% of pT1b (SM2-3) patients were treated with ESD as the primary treatment before esophagectomy. These were non-curative ESDs, and additional adjuvant esophagectomy with 2- or 3-field LN dissection was later performed. Among these patients, we selected 27, 38, and 64 ESD margin negative patients in the pT1a (MM), pT1b (SM1), and pT1b (SM2-3) groups, respectively. The NPVs in the pT1a (MM), pT1b (SM1), and pT1b (SM2-3) groups were 85.7%, 80.0%, and 78.0%, respectively (Table II, Table III, and Table IV). In the pT1a (MM) and pT1b (SM1) groups, the NPVs were beyond 85% and 80%, respectively.
Discussion
This multi-institutional observational study revealed that a new diagnostic test kit for evaluating CRP 1846C>T genetic polymorphism (rs1205) is not useful for assessing the risk of LN metastasis in pT1a (MM) or pT1b(SM1), cN0 thoracic ESCC. The NPV for LN involvement using the new test was 82.8% in pT1a (MM) ESCC, which suggests the new test has the potential to extend the indication of ESD. However, its power to diagnose the risk of LN metastasis in pT1a (MM), cN0 cancer was not sufficient for use in a clinical setting.
In an earlier report, we demonstrated that the NPV for lymph node involvement using CRP 1846C>T polymorphism was 80% in pT1b (SM) ESCC (19). The discrepancy between that earlier result and the results in the present study appear to be due to patient selection. The earlier study included cN1 patients with clinical LN metastasis, whereas the present study included only cN0 patients. In the earlier study, therefore, more than 70% of patients had lymphatic invasion and more than 40% had pathological LN metastasis. These are higher than in the present study, where the incidences among pT1b (SM2-3) patients were 28.2% and 43.7%, respectively. This shows that diagnosis of recurrence risk becomes more difficult in situations where lymph node metastasis is infrequent.
We have already reported that the test is useful for lung, breast, and uterine cancers, and we are currently testing its utility in gastric and colorectal cancer. However, these studies include cN1 patients (20, 21). Our findings so far suggest this test is effective for assessing the risk of LN metastasis in a variety of solid tumors. We hope that future studies will reveal that this risk diagnosis works more effectively for gastric and colorectal cancer than for esophageal cancer.
ER combined with adjuvant therapy appears to be a new treatment option for cT1N0M0 ESCC invading to pT1a (MM) with LVI or pT1b (22). Adjuvant esophagectomy should be the treatment modality for patients at high risk of relapse (11-13). However, considering the benefits of organ preservation, adjuvant CRT can be a treatment option for patients not at high risk. Minashi et al., reported the efficacy of the combination of ER and adjuvant selective CRT in patients with pT1b (SM1-2) N0M0 thoracic ESCC (22). The 3-year overall survival rate was 92.6% among all patients (90% confidence interval=88.5%-95.2%). The efficacy was thus comparable to that of adjuvant esophagectomy, and the combination of ER and selective CRT should be considered as a minimally invasive treatment option. However, they reported several grade 3-4 adverse events associated with CRT, including neutropenia (22.9%), hyponatremia (7.3%), and anorexia (7.3%) (19). Moreover, cardiac- and lung-associated late adverse events at grades >2 were observed in 8 (8.3%) patients, with one patient suffering grade 4 cardiac ischemia (22). No matter how selective CRT is, caution is required, as it is not as safe as ER monotherapy.
This study has several limitations. First, there was a selection bias in that patients who had died due to recurrence of their esophageal cancer (including LN recurrence) were excluded because they could not give us consent to use their data. Second, we did not clearly show the rationale behind the use of this test to assess LN metastasis risk. CRP gene is located on chromosome 1q21-q23, and although the rs1205 polymorphism is within the gene’s noncoding region (1846C>T), it is associated with differences in baseline serum CRP levels (23, 24). Patients with the CRP 1846 T/T genotype produce less CRP than those with the C/C or C/T genotype, which reduces their baseline level of CRP (22, 24). We previously reported that, in vitro, CRP suppresses murine colon adenocarcinoma cell migration and may also inhibit epithelial-to-mesenchymal transition in oral squamous cell carcinoma cells by suppressing N-cadherin and ZEB-1 expression (25). In vivo, baseline CRP appears to inhibit tumoral lymphangiogenesis and lymph node metastasis in mice (26). Furthermore, higher baseline CRP decreases the accumulation of M2 macrophages and angiogenesis within tumors (27). We recently reported that peritumoral accumulation of CD16b-positive neutrophils is an independent factor strongly correlated with LN metastasis in ESCC (28). This suggests that peritumoral inflammation influences LN metastasis, which may indicate the involvement of CRP. However, further basic study is necessary to address these issues.
In conclusion, the developed test kit for CRP 1846C>T genetic polymorphism is not useful for assessing the risk of LN metastasis in T1a-MM with LVI or T1b-SM1 ESCC with a sufficient power to extend the ESD indication clinically. However, this study also showed that the possibility that CRP gene polymorphisms are involved in the mechanism of lymph node metastasis in solid tumors still remains.
Acknowledgements
This study was supported by AMED under Grant Number JP 19lm0203079. The Authors would like to express our heartfelt gratitude to our collaborators at 65 facilities in allover Japan (B-37 CRP-SNP Study Group). The Authors would also like thank Prof. Ikeda, Ms. Fujiki, Mr. Sakagami, Ms. Ono, and Ms. Kobayashi (Clinical Research Innovation and Education Center, Tohoku University Hospital, Sendai, Japan), Mr. Kawabe for monitoring (Clinical Research Data Center, Tohoku University Hospital, Sendai, Japan), and Yosuke Sato (Akita University Graduate School of Medicine, Akita, Japan) for their assistance.
Footnotes
Authors’ Contributions
This project and the analysis were designed and conducted by Satoru Motoyama. All named Authors participated in the development of this manuscript and in the decision to submit this manuscript for publication. The Authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the accuracy and integrity of any part of the work as a whole and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved, and have given their approval for this version to be published. Satoru Motoyama: conceptualization, data curation, funding acquisition, investigation: methodology, project administration, resources, writing – original draft, editing. Miki Hosaka, Takashi Kamei, Mitsuharu Hirai, Manami Hoshi, Masatomo Miura: conceptualization, data curation, funding acquisition, methodology, project administration. Munenori Takata: data curation, methodology. Kazuma Fujita, Mie Kanda: investigation. Masaki Ueno, Masayuki Watanabe, Yu Ohkura, Shinji Mine, Yu Ohkura, Shinji Mine, Shunsuke Tanabe, Takahiro Toyokawa, Akiyuki Wakita, Katsunori Nishikawa, Itasu Ninomiya: resources (patients). Shih-wei Chiu, Takuhiro Yamaguchi: formal analysis.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
- Received October 4, 2022.
- Revision received October 15, 2022.
- Accepted October 17, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.