Prognostic impact of RAS mutation on papillary thyroid carcinoma: a systematic review and meta-analysis Authors: Alireza Abdollahi, Seyed Morteza Pourfaraji, Fatemeh Ojaghi Shirmard, Reza Ghalehtaki, Samaneh Salarvand Published: 2025 ABSTRACT Background and aimPapillary thyroid carcinoma (PTC) is the most common form of thyroid malignancy, and activating mutations in the RAS gene family have been implicated in its pathogenesis and clinical behavior. This study aimed to evaluate the role of RAS mutations in clinical outcomes of patients with PTC.MethodsWe conducted a comprehensive literature search of PubMed, Scopus, and WOS. Eligible studies included data on RAS mutation prevalence and at least one of the predefined endpoints. Hazard ratios (HR) for recurrence-free survival (RFS), odds ratios (OR) for mortality, distant metastasis, and tumor recurrence, and mean differences (MD) for tumor size were used as effect sizes.ResultsA total of 22 studies, involving 4727 patients, were included in this analysis. The pooled prevalence of RAS mutations was 20% (95% CI: 14%-28%). Presence of RAS mutation was associated with significantly higher odds of distant metastasis (OR = 3.12, 95% CI: 1.48–6.59, p < 0.01), while no significant associations were found with RFS (HR = 1.05, 95% CI: 0.52–2.13, p = 0.88), overall mortality (OR = 2.01, 95% CI: 0.60–6.70, p = 0.21), tumor recurrence (OR = 0.35; 95% CI: 0.08–1.61, p = 0.13). Moreover, our findings revealed no significant differences in tumor size between RAS-positive and RAS-negative cases (MD = 0.01, 95% CI: -0.73–0.72, p = 0.96).ConclusionsRAS mutations are present in 20% of PTC cases and are associated with increased risk of distant metastasis. However, these mutations were not significantly associated with other clinical outcomes. These findings support the integration of RAS mutation testing into postoperative risk stratification to identify patients who may benefit from more intensive surveillance and personalized therapeutic strategies. Introduction Papillary thyroid carcinoma (PTC) is the most common form of thyroid cancer, accounting for approximately 80–85% of all thyroid malignancies. In 2020, the GLOBOCAN database estimated 586,202 new cases of thyroid cancer worldwide, making it the 10th most frequently diagnosed cancer globally [1, 2]. Despite its rising incidence, PTC is associated with an excellent overall prognosis. Global mortality for thyroid cancer has remained stable over recent decades at roughly 0.5 deaths per 100,000 population [3, 4]. Standard treatment involves total or near-total thyroidectomy, often followed by radioiodine (RAI) ablation and thyroid-stimulating hormone (TSH) suppression therapy. Rising use of lobectomy for low-risk tumors, active surveillance for microcarcinomas, and molecular stratification have refined management algorithms, aiming to minimize overtreatment while maintaining favorable oncologic outcomes [5, 6]. Molecular profiling has emerged as a pivotal adjunct in the diagnosis, prognostication, and therapeutic planning of PTC, with RAS gene mutations among the most studied alterations. Activating point mutations in HRAS, KRAS, and NRAS, particularly at codons 12, 13, and 61, lead to constitutive signaling through the MAPK and PI3K/AKT pathways, driving tumorigenesis by enhancing proliferation, survival, and motility [7, 8]. In diagnostic cytopathology, RAS mutations are detected in approximately 20–30% of indeterminate thyroid nodules, which can improve risk stratification and reduce unnecessary surgical interventions when combined with other molecular markers [9–11]. Since RAS-positive PTCs often present as the follicular variant, a histological subtype that can mimic benign follicular lesions, detection of RAS mutations may aid preoperative decision-making and tailoring the extent of surgery. While the efficacy of direct RAS inhibitors remains unclear, tumor RAS status may inform consideration of downstream targeted therapies, such as MEK or PI3K inhibitors, or enrollment in clinical trials of novel agents in advanced or RAI-refractory cases [12, 13]. In the current study, we aimed to evaluate the prognostic effect of RAS mutation on PTC. Methods and materials This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Systematic search A comprehensive search strategy was conducted across major databases, including Web of Science, Scopus, and PubMed, encompassing all available records up to June 2025. The search utilized relevant Medical Subject Headings (MeSH) terms and keywords specific to RAS mutation in papillary thyroid carcinoma, focusing on (“prognosis” OR “survival” OR “recurrence” OR “hazard” OR “metastasis”) AND (“HRAS” OR “KRAS” OR “RAS mutation” OR “RAS” OR “NRAS”) AND (“papillary thyroid carcinoma” OR “PTC” OR “thyroid cancer”). Inclusion and eligibility criteria Eligibility criteria were established based on the PICO framework: Population (P): Adult patients diagnosed with PTC. Intervention (I): Detection of the RAS mutation. Comparison (C): Patients with RAS-mutated versus wild-type status. Outcome (O): Clinical outcomes, including Recurrence-free survival (RFS), tumor recurrence, distant metastasis, and mortality. In addition, studies reporting the prevalence of RAS mutations in PTC were also included. Studies were excluded if they involved animal models, case reports, cancers other than papillary thyroid carcinoma, unclear RAS mutation testing methods, or insufficient clinical outcome data. Additionally, non-clinical studies, including histologic, genetic, or in vitro research without clinical correlation, were excluded. Data extraction and outcome measurements Data extraction was performed independently by two authors using a standardized data collection form, with discrepancies resolved by consulting a third author. Extracted variables included: study authors, year of publication, study design, sample size, patient demographics, RAS mutation status, follow-up periods, and clinical outcomes including RFS, tumor recurrence, distant metastasis, mortality, and tumor sizes along with their respective hazard ratio (HR), odds ratio (OR), mean difference (MD) and corresponding confidence intervals (CI). Recurrence was defined as the reappearance of PTC following a period of remission, identified through a combination of cytology or pathology, imaging studies, and biochemical markers. Mortality data primarily reflect overall mortality, as reported by the majority of the analyzed publications. Statistical analysis and data synthesis The pooled HR for RFS, OR for tumor recurrence, distant metastasis, and mortality, and MD for tumor size in patients with PTC, were calculated using a random-effects model. Effect sizes for each outcome were estimated by pooling logarithmic transformations of HR and OR along with their corresponding 95% CIs. Heterogeneity across studies was assessed using the I² statistic to evaluate heterogeneity among results. Forest plots were generated to visually represent the pooled estimates and heterogeneity, while funnel plots were constructed to assess potential publication bias whenever more than 10 studies were included. Statistical significance was considered as a p-value < 0.05 in the current analyses. All statistical analyses were performed using the “meta” package in R (R Foundation for Statistical Computing, Vienna, Austria). Results Our initial search yielded 1,096 articles from PubMed, Scopus, and Web of Science, from which we eliminated 369 duplicates. After reviewing the titles and abstracts of the remaining 727 records, we retrieved 61 full-text articles for further evaluation. Ultimately, 22 studies [6–8, 13–31] met our eligibility criteria and were included in the systematic review and meta-analysis (Fig. 1). Detailed characteristics of the included studies are summarized in Table 1. Fig. 1PRISMA flow diagram of the included studies PRISMA flow diagram of the included studies Table 1Detailed characteristics of the included studiesAuthorYearCountrySampleSizeDeathDistant MetastasesRelapseRFSNodule DiameterRAS+RAS-RAS+RAS-RAS+RAS-HRULRAS+RAS-N (%)N (%)N (%)N (%)N (%)N (%)MeanSD N MeanSD N An [14]2015Korea155---------3.61.2564.72.956Craig [15]2023Australia491---------------Denti [16]2024Italy69---------1.90.9251.50.940Fukahori [17]2012Japan985 (15.1)0 (0.0)11 (33.3)0 (0.0)4 (15.4)1 (4.0)---------Hara [18]1994USA915 (38.4)1 (1.2)8 (61.5)0 (0.0)-----------Jang [19]2014Korea853 (7.6)6 (13)13 (33.3)15 (32.6)--1.130.112.43.21.3433.6291Kody [20]2018AUS100--1 (8.3)12 (17.1)4 (33.3)24 (34.3)---3.12.2122.31.770Jung [6]2019Korea612 (15.3)5 (10)10 (76.9)13 (27.1)-----------Kim [21]2017Korea177--1 (1.4)0 (0.0)--0.480.063.772.70.6471.80.2240Li [22]2025China96---------1.71.1691.4156Mond [23]2014Australia148--5 (71.4)3 (15.8)-----------Ozcevik [24]2025Turkey77--7 (58.3)12 (18.5)-----2.70.962.10.422Póvoa [25]2021Portugal2411(7.6)6 (3.7)0 (0.0)2 (1.3)-----------Rusinek [26]2020Poland54--3 (50.0)12 (54.5)-----------Song [7]2017Korea1341 (2.3)1 (1.1)6 (14.0)3 (3.3)2 (2.2)3 (7.0)---2.151.39231.771.6646Song [27]2016Korea5510 (0.0)2 (0.8)2 (4.3)3 (1.3)1 (2.1)21 (8.8)0.530.11.57------Staubitz [28]2023Germany170--1 (8.3)4 (2.5)-----3.42.88121.21.84158Triverdi [29]2025India74--3 (16.7)2 (4.5)1 (6.7)11 (33.3)---------Ullman [30]2022USA113------3.10.811.7------Xu [31]2020USA69--23 (79.3)27 (67.5)-----------Yip [13]2015USA15107 (2.1)18 (2.5)7 (2.2)26 (3.6)2 (0.6)60 (8.4)1.150.43.01------Zhang [8]2024China163---------1.21.8123.42.88158N = Number of cases; RFS = Recurrence-free survival; HR = Hazard ratio; U = Upper 95% confidence interval; L = Lower 95% confidence interval; SD = Standard deviation Table 1Detailed characteristics of the included studiesAuthorYearCountrySampleSizeDeathDistant MetastasesRelapseRFSNodule DiameterRAS+RAS-RAS+RAS-RAS+RAS-HRULRAS+RAS-N (%)N (%)N (%)N (%)N (%)N (%)MeanSD N MeanSD N An [14]2015Korea155---------3.61.2564.72.956Craig [15]2023Australia491---------------Denti [16]2024Italy69---------1.90.9251.50.940Fukahori [17]2012Japan985 (15.1)0 (0.0)11 (33.3)0 (0.0)4 (15.4)1 (4.0)---------Hara [18]1994USA915 (38.4)1 (1.2)8 (61.5)0 (0.0)-----------Jang [19]2014Korea853 (7.6)6 (13)13 (33.3)15 (32.6)--1.130.112.43.21.3433.6291Kody [20]2018AUS100--1 (8.3)12 (17.1)4 (33.3)24 (34.3)---3.12.2122.31.770Jung [6]2019Korea612 (15.3)5 (10)10 (76.9)13 (27.1)-----------Kim [21]2017Korea177--1 (1.4)0 (0.0)--0.480.063.772.70.6471.80.2240Li [22]2025China96---------1.71.1691.4156Mond [23]2014Australia148--5 (71.4)3 (15.8)-----------Ozcevik [24]2025Turkey77--7 (58.3)12 (18.5)-----2.70.962.10.422Póvoa [25]2021Portugal2411(7.6)6 (3.7)0 (0.0)2 (1.3)-----------Rusinek [26]2020Poland54--3 (50.0)12 (54.5)-----------Song [7]2017Korea1341 (2.3)1 (1.1)6 (14.0)3 (3.3)2 (2.2)3 (7.0)---2.151.39231.771.6646Song [27]2016Korea5510 (0.0)2 (0.8)2 (4.3)3 (1.3)1 (2.1)21 (8.8)0.530.11.57------Staubitz [28]2023Germany170--1 (8.3)4 (2.5)-----3.42.88121.21.84158Triverdi [29]2025India74--3 (16.7)2 (4.5)1 (6.7)11 (33.3)---------Ullman [30]2022USA113------3.10.811.7------Xu [31]2020USA69--23 (79.3)27 (67.5)-----------Yip [13]2015USA15107 (2.1)18 (2.5)7 (2.2)26 (3.6)2 (0.6)60 (8.4)1.150.43.01------Zhang [8]2024China163---------1.21.8123.42.88158N = Number of cases; RFS = Recurrence-free survival; HR = Hazard ratio; U = Upper 95% confidence interval; L = Lower 95% confidence interval; SD = Standard deviation Detailed characteristics of the included studies N = Number of cases; RFS = Recurrence-free survival; HR = Hazard ratio; U = Upper 95% confidence interval; L = Lower 95% confidence interval; SD = Standard deviation Prevalence of RAS mutation positivity A total of 21 studies comprising 4564 cases assessed the prevalence of RAS mutation positivity in patients with papillary thyroid carcinoma. The pooled prevalence of RAS mutations was 20% (95% CI: 14%-28%; I² = 94.5%), as demonstrated by a significant heterogeneity across studies (p < 0.01) (Fig 2). Fig. 2The pooled proportion of RAS mutation positivity among PTC patients The pooled proportion of RAS mutation positivity among PTC patients Recurrent-free survival Analysis of five studies evaluating RFS included a total of 921 cases. The pooled HR using a random-effects model indicated no significant impact of RAS mutations on RFS (HR = 1.05, 95% CI: 0.52–2.13, p = 0.88). Between-study heterogeneity was low (I² = 14.2%, p = 0.32) (Fig 3). Fig. 3The pooled OR of recurrent-free survival among PTC patients with RAS mutation The pooled OR of recurrent-free survival among PTC patients with RAS mutation Mortality Eight studies assessed mortality associated with RAS mutations, encompassing a total of 1927 patients. The pooled OR for overall mortality was 2.01 (95% CI: 0.60–6.70, p = 0.21), indicating no statistically significant increase in mortality risk associated with RAS mutations. Moderate heterogeneity was present among included studies (I² = 50.2%, p = 0.05) (Fig 4). Fig. 4The pooled OR of mortality among PTC patients with RAS mutation The pooled OR of mortality among PTC patients with RAS mutation Distant metastasis Sixteen studies evaluated the association between RAS mutation and the presence of distant metastasis among 2566 cases. The pooled OR showed a statistically significant increase in distant metastasis risk in patients harboring RAS mutations (OR = 3.12, 95% CI: 1.48–6.59, p < 0.01). Significant heterogeneity was observed (I² = 61.3%, p < 0.01) (Fig 5). Fig. 5The pooled OR of distant metastasis among PTC patients with RAS mutation The pooled OR of distant metastasis among PTC patients with RAS mutation Tumor recurrence Six studies involving 1,641 patients examined the risk of tumor recurrences. The pooled OR was 0.35 (95% CI: 0.08–1.61, p = 0.13), representing a non-significant reduced odds of tumor recurrences in patients with RAS mutations, while heterogeneity was high (I² = 61.8%, p = 0.02) (Fig 6). Fig. 6The pooled OR of tumor recurrence among PTC patients with RAS mutation The pooled OR of tumor recurrence among PTC patients with RAS mutation Tumor size difference Nine studies comprising 1072 cases compared tumor size between RAS-positive and RAS-negative patients. The pooled mean difference (MD) using a random-effects model was 0.01 cm (95% CI: -0.73–0.72, p = 0.96), indicating no significant difference in tumor size between groups. Marked heterogeneity was noted (I² = 88.2%, p < 0.01) (Fig 7). Fig. 7The pooled mean difference of tumor size recurrence among PTC patients with or without RAS mutation The pooled mean difference of tumor size recurrence among PTC patients with or without RAS mutation Discussion This study revealed that approximately 20% of PTC cases harbor RAS mutations, which significantly correlate with an increased risk of distant metastasis. However, no significant associations were observed between RAS mutation positivity and RFS, overall mortality, tumor recurrence, or tumor size. These findings support the prognostic importance of RAS mutations predominantly in the context of distant metastasis, suggesting a nuanced role of these mutations in the clinical trajectory of PTC [5, 10, 32, 33]. Our study findings on the prevalence of RAS mutations (20%) align closely with prior research, which reported mutation frequencies ranging from 10% to 20%. This consistency reinforces the robustness of RAS mutation prevalence data across different populations and methodologies [34, 35]. However, high heterogeneity observed across studies indicates potential variability due to geographic, demographic, or methodological differences. Previous research frequently highlighted significant associations between RAS mutations and aggressive clinical outcomes, such as an increased rate of tumor distant metastasis. Our analysis confirmed this relationship, demonstrating a significantly elevated risk of distant metastasis among patients harboring RAS mutations. The increased recurrence risk associated with RAS mutations underscores their potential as a marker for identifying patients who might benefit from more vigilant monitoring, and potentially adjunctive treatments [8, 33, 36]. In contrast, our analysis found no significant relationship between RAS mutations and mortality or tumor recurrence, differing from some prior studies that indicated potential associations with worse prognosis. Earlier studies, notably those with smaller sample sizes or single-institution designs, have occasionally reported a significant correlation between RAS mutations and poorer survival outcomes [28, 37]. This discrepancy may be attributed to several factors, including variations in sample size, statistical power, methodological rigor, follow-up duration, and differences in patient treatment protocols across studies. Specifically, the variability in follow-up periods can significantly influence the detection of mortality and metastasis outcomes, as thyroid carcinoma typically demonstrates indolent progression, potentially delaying the identification of long-term prognostic impacts [38, 39]. RAS mutations play a crucial role in the molecular pathogenesis of PTC, influencing tumor behavior through several intricate biological pathways. These mutations commonly affect the HRAS, KRAS, and NRAS genes, integral components of the MAPK (mitogen-activated protein kinase) and PI3K/AKT signaling pathways. The activation of these pathways leads to altered cellular processes including proliferation, differentiation, apoptosis, and migration, significantly contributing to tumorigenesis and cancer progression in PTC [11, 40]. The activation of RAS proteins typically occurs through point mutations in codons 12, 13, and 61, causing a constitutive activation of downstream signaling pathways. This continuous signaling results in increased cellular proliferation and survival, fostering a tumorigenic environment. Specifically, activation of the MAPK pathway promotes uncontrolled cell division and growth, whereas the PI3K/AKT pathway primarily enhances cellular survival and metabolic reprogramming, supporting sustained tumor growth [29, 41, 42]. Clinically, the presence of RAS mutations in PTC patients carries important prognostic implications. While PTC is generally associated with an excellent prognosis, subsets harboring RAS mutations may present with distinct clinical characteristics [43, 44]. Patients with RAS-positive tumors frequently demonstrate a follicular variant of PTC, characterized histologically by follicular architecture, which has implications for both diagnosis and clinical management. Due to the histological similarities between follicular-variant PTC and benign follicular lesions, the identification of RAS mutations can serve as a valuable diagnostic tool to enhance diagnostic accuracy and facilitate optimal therapeutic decisions [12, 22]. The clinical significance of RAS mutations extends beyond distant metastasis. Despite the clear association with distant metastasis, our analysis revealed no significant impact on mortality, suggesting a complex, nuanced role for RAS mutations in PTC prognosis [32, 45]. This complexity may arise from the interaction between RAS-driven signaling and other genetic and epigenetic alterations present in tumor cells, influencing clinical behavior in a multifactorial manner. Interestingly, the inconsistent findings regarding the correlation of RAS mutations with tumor size highlight another layer of complexity. Previous studies have variably reported RAS mutations in association with larger tumors or advanced pathological stages, while other studies, including our analysis, found no significant correlation [10, 46, 47]. The considerable heterogeneity observed in the tumor size analysis (I² = 88%) might explain these conflicting results. Further, these discrepancies suggest that tumor growth regulation in RAS-mutated PTC involves additional factors beyond simple RAS activation, potentially including interactions with the tumor microenvironment, immune responses, and coexisting molecular alterations. From a therapeutic perspective, understanding the molecular underpinnings of RAS mutations presents opportunities for targeted therapies. However, the clinical application of RAS-targeted therapies in PTC remains challenging due to the absence of effective direct RAS inhibitors and the complexity of downstream pathways [35, 48, 49]. Nonetheless, the identification of RAS mutations may guide personalized treatment approaches, including closer surveillance and consideration of targeted or adjuvant therapies aimed at downstream effectors of the RAS signaling pathway. This study has several limitations. The considerable heterogeneity observed across the included studies, possibly due to variations in patient demographics, diagnostic methods, molecular testing approaches, and follow-up durations, may limit the generalizability of the findings. Additionally, the observational nature of the studies analyzed could introduce potential selection and reporting biases. Furthermore, limited reporting of detailed histopathological features and presurgical clinical data in the included studies restricted our ability to interpret these prognostic factors. Lastly, the lack of standardized reporting in tumor characteristics and outcomes across studies emphasizes the need for uniform methodologies in future research to confirm and expand upon these findings. Conclusions In conclusion, our systematic review and meta-analysis affirm the prognostic importance of RAS mutations in PTC, particularly their role in predicting distant metastasis. While RAS mutations did not significantly impact mortality, recurrences, or tumor size, their presence clearly aids in refining clinical management strategies through enhanced diagnostic accuracy and tailored therapeutic approaches. Future investigations incorporating standardized methodologies and extended follow-up periods will be crucial in further elucidating the full prognostic and therapeutic implications of RAS mutations and individual RAS subtypes (KRAS, NRAS, HRAS) in PTC.