A nomogram for predicting overall survival in patients with gastric cancer based on tumor suppressor RCAN1.4 expression and clinical risk factors Authors: Qiang Zhao, Congxiu Miao, Qingpu Lu, Weipeng Wu, Yuan He, Mengzhu Wang, Huimin Liu, Jiangman Zhao, Changhong Lian Published: 2024 ABSTRACT Gastric cancer (GC) is one of the most prevalent malignant tumors in the world and has an extremely poor prognosis. Regulator of calcineurin 1 (RCAN1), a known tumor suppressor in various cancers, has an undefined role in the proliferation and metastasis of GC. Primary tumor and paired normal gastric tissues were collected from 77 patients with GC for evaluating the mRNA levels of 3 RCAN1 transcripts. Kaplan–Meier survival curves and Cox regression analysis were used to assess the prognostic value of 3 RCAN1 transcripts, and to select variables for nomogram. The mRNA levels of RCAN1 isoform 1 (RCAN1.1, P = .0312) and isoform 2 (RCAN1.2, P = .007) were significantly diminished in GC tissues compared with normal tissues, whereas isoform 4 (RCAN1.4) expression level showed no significant differences. GC patients with lower RCAN1.4 mRNA levels had shorter overall survival time than patients whose tumors had high RCAN1.4 levels (P = .04). Downregulated expression of RCAN1.4 was found to be an independent prognostic factor of overall survival in GC patients, with a hazard ratio of 2.485 and a significant P-value of .023 in multivariate Cox analysis. The concordance index of nomogram to predict overall survival was 0.788, based on RCAN1.4 level, tumor stage and lymph node metastasis status. In conclusion, our findings suggest that RCAN1.4 is a novel prognostic marker for gastric cancer, targeting RCAN1.4 may provide a promising therapeutic strategy in GC management. 1. Introduction Regulator of calcineurin 1 (RCAN1) gene, located on chromosome 21, is comprised of 7 exons and 6 introns. Three transcripts (RCAN1.1, RCAN1.2, and RCAN1.4) are formed through alternative splicing in human tissues.[1] Both deficiency and overexpression of RCAN1 have been reported to promote circadian disruptions in conditions such as Down syndrome, Alzheimer disease, and possibly aging.[2] Recently, studies have clearly revealed that RCAN1 plays an important role in the development of various cancers. For example, it may act as a tumor suppressor gene in small cell lung cancer.[3] Especially, RCAN1.4 has been implicated as a growth and metastasis suppressor for thyroid cancer[4] and has been shown to hinder nuclear translocation of NFAT1, thereby acting as a tumor suppressor in hepatocellular carcinoma.[5] Given these observations, RCAN1-mediated calcineurin inhibition had been considered a potential therapeutic target for cancer.[6] However, the roles of RCAN1 in gastric cancer are yet to be elucidated. Gastric cancer (GC) ranked amongst the top 5 cancers diagnosed and cancer-related mortality in 2022 in China.[7] Unfortunately, the majority of GC patients are diagnosed at advanced stages, with the overall 5-year survival rate of these patients remaining very low (<10%).[8,9] To reduce mortality, numerous biomarkers have been explored to facilitate early GC detection, elucidate underlying mechanisms of GC, and improve the prognosis estimates.[10–14] As reported by Razvan Cristescu et al, GC patients with microsatellite-unstable tumor exhibit the most favorable prognosis, and patients with microsatellite-stable and TP53-active tumor showed better prognosis than that with TP53-inactive tumor.[10] An array of miRNAs, including miR-21, miR-106a,[15] miR-421[16] and so on, were found to be overexpressed in GC tissues and patients’ serum/plasma compared to normal controls, indicating their potential as early screening biomarkers of GC. However, the role of several miRNAs in GC remains to be thoroughly investigated.[11] Gastric tumorigenesis is a complicated process, characterized as polyfactorial, multicellular, and multistep.[17] External triggers such as diet,[18] exogenous chemicals,[19] Helicobacter pylori,[20] and intragastric formation of carcinogens can induce the GC development. Furthermore, the formation and progression of GC are influenced by the interplay between GC cells and tumor microenvironment cells.[16–18] Despite these insights, the molecular mechanisms of GC initiation and progression are still not fully understood.[21,22] Further studies are needed to learn more about the pathologic mechanisms of GC and to develop new methods for diagnosis and treatment. This study aims to determine the specific role of the RCAN1 gene in the growth and metastasis of gastric cancer. Using 77 pairs of gastric cancer tissues and paired nontumor tissues, we performed real-time polymerase chain reaction (RT-PCR) analyses to quantify the mRNA levels of 3 RCAN1 transcripts. To our knowledge, we present the first demonstration of a specific tumor suppressive function of RCAN1 in the context of gastric cancer. These results provide evidence for a critical role of the RCAN1 gene during the progression of gastric cancer. 2. Methods 2.1. Patients and samples We recruited 77 primary gastric cancer patients who underwent curative resection at the Heping Hospital (Changzhi Medical College, Shanxi, China) between March 2018 and March 2019. Freshly frozen tumor tissues and paired normal gastric tissues were collected for real-time PCR. Histological confirmation of these samples was carried out by 2 independent pathologists. None of the patients had received preoperative radiotherapy or chemotherapy. This study was approved by the Medical Ethics Committee of Heping Hospital (2023026), Changzhi Medical College. Written informed consent was obtained from each patient. 2.2. Clinical information and follow up The demographic and clinicopathological characteristics of the entire study cohort are summarized in Table 1. After hospital discharge, the patients were followed-up every 3 months during the initial 2 years, and every 6 months from the third to the fifth year. Overall survival (OS) was defined as the time from surgery to death. During the follow-up period, 39 (50%) of 78 patients reached the follow-up endpoint (death), the other 39 patients still being followed up. Table 1Correlation between RCAN1.1 expression and different clinicopathological features of gastric cancer patients.CharacteristicsTotalRCAN1.1 expressionP-valueUpDownGender.096 Male592633 Female18414Age (years).012* ≤60351916 >60421131BMI (kg/m2).820 <18.5945 18.5–23.9401426 ≥24271116Smoking.668 Yes21912 No562135Drinking.641 Yes422 No732845Stage.893 I–II241014 III–IV451827Differentiation.382 Poor411724 Moderate23716Location.742 Gastric cardia391821 Gastric body23914 Gastric angle514 Gastric antrum16610Tumor size.189 <5 cm381226 ≥5 cm251213Lymph node metastasis.333 Yes391821 No291019Up, RCAN1.1 expression was upregulated in tumor tissue compared with adjacent normal tissue; Down, RCAN1.1 expression was downregulated in tumor tissue compared with adjacent normal tissue.BMI = body mass index.*P < .05. The cases missing corresponding information were not presented. Correlation between RCAN1.1 expression and different clinicopathological features of gastric cancer patients. Up, RCAN1.1 expression was upregulated in tumor tissue compared with adjacent normal tissue; Down, RCAN1.1 expression was downregulated in tumor tissue compared with adjacent normal tissue. BMI = body mass index. P < .05. The cases missing corresponding information were not presented. 2.3. RNA extraction and quality control We extracted total RNA from 77 paired samples of gastric cancer and adjacent noncancerous gastric tissues using TRIzol reagent (Invitrogen, Carlsbad, CA) as per the manufacturer’s protocol. RNA concentration and quality were measured by a NanoDrop ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE). Electropherograms of the RNA samples were obtained by an Agilent 2100 Bioanalyzer platform (Agilent Technologies, Santa Clara, CA). The RNA integrity number was calculated using Agilent 2100 Expert software version B.02.03.SI307. The quality control criteria of RNA samples included (1) concentration ≥ 30 ng/uL; (2) the 260/280 nm absorbance ratio was 1.8 to 2.0. 2.4. RT-PCR assay of 3 RCAN1 isoforms Following the manufacturer’s instruction, the reverse transcription was conducted using the PrimeScriptTM RT Reagent Kit (Takara, Dalian, China), and the RT-PCR was immediately carried out with SYBR Premix Ex Tag (Takara, Dalian, China). RCAN1.1 was amplified using forward primer: 5′-TGGAGCTTCATTGACTGCGA-3′, and reverse primer: 5′-ACGTCCTAAAGAGGGACTCA-3′. RCAN1.2 was amplified using forward primer: 5′-GGCGACGTGACTCAGTGTTC-3′, and reverse primer: 5′-GGTGATGTCCTTGTCATACGTC-3′. RCAN1.4 was amplified using forward primer: 5′-TTTAGCTCCCTGATTGCCTGT-3′, and reverse primer: 5′-AAAGGTGATGTCCTTGTCATACG-3′. The quantitative PCR reaction was performed as following: 95 °C 15 seconds for denaturation, then 40 cycles of 95 °C for 5 seconds and 60 °C for 30 seconds, with all samples in duplicates. The gene expression levels were calculated and normalized against the internal reference gene GAPDH, and the comparative CT method was used for comparisons between samples. The relative expression levels were calculated by the equation: 2‐ΔCT (ΔCT = CTTarget ‐ CTGAPDH). When analyzing the relationship between the expression levels of RCAN1 transcripts and clinical features, the Log22‐ΔΔCT (ΔΔCT = ΔCTCancerous tissue ‐ ΔCTNoncancerous tissue) value >0 indicated that RCAN1 transcript expression was upregulated in tumor tissue compared with paired normal tissue (the Up group), and <0 meant that RCAN1 transcript expression was downregulated in tumor tissue compared with paired normal tissue (the Down group), which divided this GC cohort into 2 groups. 2.5. Statistical analysis Statistical analysis was performed using SPSS 22 program (SPSS Inc., Chicago, IL), GraphPad Prism 7 (GraphPad Software, La Jolla, CA), and R project. Wilcoxon matched-pairs signed rank test were used to analyze the differences of 3 RCAN1 isoforms’ mRNA expression level between tumor tissues and matched normal tissues. Mann–Whitney test was used to analyze the difference of 3 RCAN1 isoforms’ mRNA relative expression level between 2 subgroups of GC patients. The statistic power achieved 0.988 when comparing the level of RCAN1.1, RCAN1.2, and RCAN1.4 between tumor and normal tissues from 77 patients using Wilcoxon matched-pairs signed rank test of G*Power 3.1.9.7. Chi-square test or Fisher exact test was performed in the comparison of the clinical characteristics within 2 groups, and Kaplan–Meier analysis was used for the survival assessment. Log-rank test was conducted to compare the survival rate of patients between subgroups. A nomogram was constructed based on prognostic factors associated with OS identified by Cox proportional hazards model. Internal validation by the bootstrap method was performed, the concordance index (C-index) and calibration curve were calculated to assess the predictive accuracy and discriminatory capacity. The nomogram was generated using R 4.0.2 and third-party package rms and its auxiliary program package. Difference was marked as statistically significant with *P value < .05, **P < .01, and *** P < .001. 3. Results 3.1. mRNA levels of 3 RCAN1 transcripts in paired GC samples Real-time PCR were performed to detect the mRNA levels of 3 RCAN1 transcripts in both GC tissues and paired noncancer gastric tissues. Transcript RCAN1.1 (P = .0312) and RCAN1.2 (P = .007) were significantly decreased in GC tissues compared to their corresponding noncancer gastric tissues (Fig. 1A and B). However, mRNA level of RCAN1.4 exhibited no significant difference in paired GC samples (P = .1472, Fig. 1C). To further explore the correlation among 3 transcripts of RCAN1 in GC tissues, Spearman correlation analysis was performed. The results showed there was positively and significantly correlation between any 2 transcripts (Fig. 1D). The mRNA levels of 3 RCAN1 transcripts in 77 pairs of GC samples were determined by real-time PCR. GAPDH was used as a loading control. Scatter plot of RCAN1.1 (A), RCAN1.2 (B), and RCAN1.4 (C). The mRNA levels in paired GC samples. (D) Heatmap of correlation among RCAN1.1, RCAN1.2, and RCAN1.4 relative expression levels in tumor tissues of gastric cancer patients. The value in grids of the upper triangle is Spearman correlation coefficient (Spearman r). The value in grids of the lower triangle is P value of Spearman correlation. ***P < .001. 3.2. Correlation between 3 RCAN1 transcripts’ mRNA expression and clinicopathological characteristics of GC patients To investigate the relationship between the expression levels of RCAN1 transcripts and clinical characteristics, we divided our GC cohort into 2 groups according to the value of Log22‐ΔΔCT as described in the methods. RCAN1 transcript expression was upregulated in tumor tissue compared with adjacent normal tissue in the Up group, while downregulated in the Down group. RCAN1.1 mRNA level was downregulated in 61% (47/77, Table 1) of all cases of gastric cancer patients. Comparatively, 58% (45/77, Table 2) patients showed downregulated RCAN1.2 mRNA level, and 48 of 77 (62%, Table 3) patients were in the Down group of RCAN1.4. Table 2Correlation between RCAN1.2 expression and different clinicopathological features of gastric cancer patients.CharacteristicsTotalRCAN1.2 expressionP-valueUpDownGender.793 Male592534 Female18711Age.109 ≤60351817 >60421428BMI.244 <18.5954 18.5–23.9401921 ≥2427819Smoking.238 Yes211110 No562135Drinking.637 Yes413 No733142Stage.245 I–II241212 III–IV451629Differentiation.380 Poor411526 Moderate231112Location.560 Gastric cardia392019 Gastric body231112 Gastric angle514 Gastric antrum16610Tumor size.328 <5 cm381523 ≥5 cm251312Lymph node metastasis.498 Yes391821 No291118Up, RCAN1.2 expression was upregulated in tumor tissue compared with adjacent normal tissue; Down, RCAN1.2 expression was downregulated in tumor tissue compared with adjacent normal tissue.BMI = body mass index.The cases missing corresponding information were not presented. Correlation between RCAN1.2 expression and different clinicopathological features of gastric cancer patients. Up, RCAN1.2 expression was upregulated in tumor tissue compared with adjacent normal tissue; Down, RCAN1.2 expression was downregulated in tumor tissue compared with adjacent normal tissue. BMI = body mass index. The cases missing corresponding information were not presented. Table 3Correlation between RCAN1.4 expression and different clinicopathological features of gastric cancer patients.CharacteristicsTotalRCAN1.4 expressionP-valueUpDownGender.498 Male592138 Female18810Age.023* ≤60351817 >60421131BMI.917 <18.5945 18.5–23.9401525 ≥24271017Smoking.962 Yes21813 No562135Drinking.999 Yes413 No732845Stage.982 I–II24915 III–IV451728Differentiation.033* Poor412021 Moderate23518Location.968 Gastric cardia391425 Gastric body23914 Gastric angle523 Gastric antrum1679Tumor size.001** <5 cm38929 ≥5 cm25169Lymph node metastasis.292 Yes391722 No29920Up, RCAN1.4 expression was upregulated in tumor tissue compared with adjacent normal tissue; Down, RCAN1.4 expression was downregulated in tumor tissue compared with adjacent normal tissue.BMI = body mass index.*P < .05, **P < .01. The cases missing corresponding information were not presented. Correlation between RCAN1.4 expression and different clinicopathological features of gastric cancer patients. Up, RCAN1.4 expression was upregulated in tumor tissue compared with adjacent normal tissue; Down, RCAN1.4 expression was downregulated in tumor tissue compared with adjacent normal tissue. BMI = body mass index. P < .05, **P < .01. The cases missing corresponding information were not presented. We found downregulated RCAN1.1 (P = .012, Table 1) and RCAN 1.4 (P = .023, Table 3) levels were significantly related with elder age. A similar phenomenon was also found between RCAN1.2 levels and patients’ age, though the difference was not significantly (P = .109, Table 2). Surprisingly, downregulated RCAN 1.4 is significantly correlated with smaller tumor size (P = .001, Table 3) and moderate differentiation (P = .033, Table 3). There was no significant correlation between RCAN1.1, RCAN1.2, RCAN1.4 levels and patents’ other clinicopathological features such as tumor stage and lymph node metastasis status. We also performed nonparametric Mann–Whitney test to investigate the relationship between 3 RCAN1 transcripts’ level and clinicopathological factors (Figs. 2A–F and 3A–F), treating expression levels as continuous variable (Log22‐ΔΔCT). We found both downregulated RCAN1.1 (P = .0431, Fig. 2D) and RCAN1.4 (P = .0216, Fig. 2F) were significantly related with smaller tumor size. However, no significance was found between RCAN1.4 levels and tumor differentiation (P = .2628, Fig. 3F), based on continuous variable. Comparison of the relative mRNA expression levels of RCAN1.1, RCAN1.2, and RCAN1.4 according to tumor stage (A–C) and tumor size (D–F) in 77 GC tissues. Line and error bar represent mean ± SD. Comparison of the relative mRNA expression levels of RCAN1.1, RCAN1.2, and RCAN1.4 according to lymph node metastasis status (A–C) and differentiation (D–F). 3.3. RCAN1.4 performed as a candidate prognostic marker for gastric cancer Previous studies have found that RCAN1 exerted tumor suppressive function in many types of solid tumors, including hepatocellular carcinoma, thyroid cancer, and small cell lung cancer. To better understand the biological function of the above-described dysregulation of RCAN1 transcripts at mRNA levels, we conducted a Kaplan–Meier survival analysis using the follow-up data from March 2018 to February 2022. Interestingly, while the mRNA levels of RCAN1.1 and RCAN1.2 had no significant effect on the OS in GC patients (Fig. 4A and B), patients with downregulated RCAN1.4 mRNA levels had shorter OS than those with upregulated RCAN1.4 mRNA levels in tumor tissues (Fig. 4C, P = .04). For clinicopathological features (Fig. 5A–F), as expected, GC patients with advanced tumor stage (Fig. 5C, P < .0001) or lymph node metastasis (Fig. 5D, P = .00024) experienced shorter OS. The univariate Cox analysis identified downregulated RCAN1.4 (Table 4, hazard ratio = 2.063, P = .042), advance tumor stage (Table 4, hazard ratio = 8.549, P < .0001) and lymph node metastasis (LNM) were risk factors of poor prognosis, while multivariate analysis identified RCAN1.4 (Table 4, hazard ratio = 2.485, P = .023) and advance tumor stage (Table 4, hazard ratio = 5.563, P = .012) as independent risk factors of poor prognosis. Stratified Kaplan–Meier analysis with log-rank test was used to visually assess the mutual effect on patient’s prognosis between RCAN1.4 level and tumor stage/LNM status (Fig. 6). The results indicated that RCAN1.4 level could furtherly refine patients’ prognosis based on tumor stage (Fig. 6A, P < .0001) and LNM status (Fig. 6B, P < .0001). Table 4Univariate and multivariate Cox analysis of factors for the overall survival of gastric cancer.VariablesUnivariate analysisMultivariate analysisHR (95%CI)P-valueHR (95%CI)P-valueAge (>60 vs ≤60 years)2.2332 (1.146–4.351).018*1.614 (0.774–3.365).202Gender (male vs female)1.152 (0.547–2.428).709RCAN1.1 (down vs up)1.832 (0.9108–3.685).0895RCAN1.2 (down vs up)1.492 (0.7746–2.872).232RCAN1.4 (down vs up)2.063 (1.025–4.151).0422.485 (1.133–5.450).023Stage (III–IV vs I–II)8.549 (2.599–28.13)<.00015.563 (1.449–21.362).012LNM (yes vs no)4.218 (1.829–9.727)<.00012.344 (0.897–6.1240.082Tumor size (≥5 vs <5 cm)1.847 (0.950–3.590).070Differentiation (moderate vs poor)1.847 (0.929–3.672).08*P < .05, ***P < .001. Univariate and multivariate Cox analysis of factors for the overall survival of gastric cancer. P < .05, ***P < .001. Kaplan–Meier overall survival curve for patients with gastric cancer between the 2 groups of RCAN1.1 up and RCAN1.1 down expression (A), RCAN1.2 up and RCAN1.2 down expression (B), RCAN1.4 up and RCAN1.4 down expression (C). Kaplan–Meier overall survival curve for patients with gastric cancer according to age (A), gender (B), tumor stage (C), lymph node metastasis status (D), tumor size (E), and differentiation (F). (A) Kaplan–Meier overall survival curve for patients with gastric cancer according to tumor stage and RCAN1.4 mRNA level. (B) Kaplan–Meier overall survival curve for patients with gastric cancer according to LNM status and RCAN1.4 mRNA level. 3.4. A nomogram to predict survival probability of patients with gastric cancer A nomogram was established based on Cox regression model with R project. Three risk factors, including RCAN1.4 level, tumor stage, and LNM status, were input for nomogram construction, which was identified by above Cox analysis results. A total score was calculated based on the point of individual factor to predict 1-year, 3-year, and 5-year survival probability of GC patients, which was showed in Figure 7A. The C-index was employed to estimate the predictive performance of nomogram, which was 0.788. The calibration curves of the nomogram were showed in Figure 7B–D, which showed good concordance between the predicted and observed values of 3-years and 5-years. (A) A nomogram model predicting 1-year, 3-year, and 5-year survival probability based on RCAN1.4 expression level, tumor stage, and lymph node metastasis status. (B–D) The calibration curves for the nomogram to predict 1-year survival probability (B), 2-year survival probability (C), and 5-year survival probability (D). 4. Discussion Due to a lack of specific symptoms in the early stages of GC, most GC patients are diagnosed at the advanced stages with poor prognosis. Consequently, it is urgent to learn more about the pathologic mechanisms and to explore more effective methods for GC diagnosis and prognosis prediction. RCAN1 exerts tumor suppressive function in many types of cancer.[4,5] However, the function of RCAN1 gene in GC progression is still obscure. In this study, we investigated the role of RCAN1 in the development of gastric cancer. RCAN1 is a gene with multiple transcriptional start sites located on chromosome 21. Three transcripts (RCAN1.1, RCAN1.2, and RCAN1.4) with alternative first exons (exon1, exon 2, and exon 4) are generated via alternatively splicing and detected in human tissues. In this study, we collected 77 pairs of GC samples and performed RT-PCR to detect the mRNA levels of 3 RCAN1 transcripts. We found that mRNA levels of RCAN1.1 (P = .0312) and RCAN1.2 (P = .0070) were significantly decreased in GC tissues compared with their corresponding noncancer gastric tissues (Fig. 1A and B). Nevertheless, RCAN1.1 and RCAN1.2 had no significant influence on patients’ prognosis. RCAN 1.1 is a major transcript highly expressed in brain,[23] exon 1 of RCAN1 has 2 translation start codons, a long and short variant corresponding to RCAN1-1L and RCAN1-1S.[24] As reported by Gennady Ermak, overexpression of RAN1-1L may promote neurodegeneration and Alzheimer disease.[25] The upregulation of RCAN1.1 in acute ischemic stroke promoted neuronal apoptosis, which could be reversed by its RNA aptamer R1SR13.[26] G. Jonatan Saenz study suggested that RCAN1.1 is a downstream target of E4BP4, which was associated with glucocorticoids-evoked apoptosis in human leukemia cells.[27] In Silvia Porta study,[28] RCAN1.2 was reported to be regulated by the Cn-activated nuclear factor of activated T-cells (NFAT) family of transcription factors. In a collaborative study, Haijun Yang et al[29] reported that low RCAN1.2 mRNA expression was related with poor prognosis of patients with esophageal squamous cell carcinoma. However, the knowledge of biological function of RCAN1.1 and RCAN1.2 in tumor development are limited explored. Though RCAN1.4 mRNA level was downregulated in 62% (48/77) of our gastric cancer patients, its expression level showed no significant difference (P = .1472) in 77 pairs of GC samples (Fig. 1C). As a competitive inhibitor for the phosphatase calcineurin, RCAN1.4 suppresses calcineurin-mediated dephosphorylation of NFAT.[30] Further studies are required to verify whether the aberrant expression of RCAN1 transcripts contributes to unique pathogenetic mechanism of gastric cancer. Our analysis revealed a significant correlation between patients’ age and RCAN1.4 expression level, with older patients or those with downregulated RCAN1.4 demonstrating shorter OS. Multivariate Cox analysis identified downregulated RCAN1.4 and advanced tumor stage as independent risk factors for poor prognosis, though there was no significant relationship between RCAN1.4 expression levels and tumor stage. To our surprise, downregulated RCAN1.4 was clearly correlated with smaller tumor size. These results indicated RCAN1.4 operates independently of traditional prognostic indicators like tumor stage. The role and mechanism of RCAN1.4 in various cancers have been well studied. For example, Rong Deng et al[31] reported RCAN1.4 played a critical role in the progression and metastasis of breast cancer by blocking CaN/NFATc1 signaling. Haojie Jin study[5] showed that RCAN1.4 in Hepatocellular Carcinoma prevents proliferation, migration, and invasive activity of cancer cells by inhibiting calcineurin activity and nuclear translocation of NFAT1. Furthermore, Bao Huang demonstrated that RCAN1.4 suppresses the osteosarcoma growth and metastasis via interfering with the calcineurin/NFAT signaling pathway.[32] RCAN1.4-calcineurin/NFAT signaling pathway was reported to be essential for hypoxic adaption of intervertebral.[30] RCAN1.4 also was reported as growth and metastasis suppressor of thyroid cancer,[4] clear cell renal cell carcinoma,[33] and pancreatic ductal adenocarcinoma.[34] To our knowledge, this is the first study that specifically identifies RCAN1.4 as a tumor suppressor in gastric cancer. Nomograms are widely used for cancer prognosis, which not only visually display the indicators that affect patients’ outcome, but also predict the survival probability by a graphical representation.[35,36] In this study, we selected the risk factors influencing the prognosis of GC patients through Cox regression analyses. The results showed that the model has high degrees of discrimination and calibrated accuracy on predicting 3-year and 5-year survival probability. However, there are still some limitations in this study. Firstly, the nomogram model lacked external validation cohort because of small sample size. Furthermore, the mechanism of RCAN1.4 contributing to regulate gastric cancer cell proliferation and growth should be elucidated and explored in future studies. 5. Conclusions Taken together, our findings suggest that RCAN1.4 serves as a candidate prognostic marker for gastric cancer. A nomogram model based on RNCA1.4 level, tumor stage and LNM status could effectively predict patients’ survival probability. Author contributions Conceptualization: Qiang Zhao, Changhong Lian. Data curation: Congxiu Miao, Qingpu Lu, Weipeng Wu, Yuan He. Formal analysis: Jiangman Zhao. Funding acquisition: Changhong Lian. Methodology: Mengzhu Wang, Huimin Liu. Project administration: Qiang Zhao. Writing – review & editing: Qiang Zhao, Jiangman Zhao, Changhong Lian. Writing – original draft: Jiangman Zhao.