Upregulation of miR-32 in tissue and plasma and its association with PTEN expression in colorectal cancer Authors: Bayan Mehrjoei, Susan Tahmasebi, Lida Haghnazari, Parmida Rezvani, Homayoon Bashiri, Nayebali Rezvani Published: 2026 ABSTRACT AbstractBackgroundEarly detection of colorectal cancer (CRC) is critical for improving patient outcomes. MicroRNAs (miRNAs) function as post-transcriptional regulators and have emerged as potential minimally invasive diagnostic biomarkers. Although miR-32 has been implicated in tumor progression in several malignancies, its expression pattern in CRC tissue and plasma and its association with PTEN expression require further clarification.Materials and methodsExpression levels of miR-32 and PTEN were measured using quantitative real-time PCR (qRT-PCR) in tumor tissues, adjacent normal tissues, and plasma samples from 55 CRC patients and plasma samples from 55 age and sex matched healthy controls.ResultsMiR-32 expression was significantly upregulated in CRC tissues and plasma samples, whereas PTEN expression was significantly reduced in tumor tissues (P < 0.001). ROC curve analysis demonstrated strong diagnostic performance of miR-32 in tissue (AUC = 0.942) and good performance in plasma (AUC = 0.846). An inverse association between miR-32 and PTEN expression was observed. Higher miR-32 levels were also associated with advanced tumor stage.ConclusionsThis is the first report of concurrent miR-32 upregulation and PTEN downregulation in both CRC tissue and plasma. MiR-32 may represent a potential non-invasive biomarker candidate for CRC detection and disease stratification. However, these findings are observational and require validation in larger, multicenter cohorts and functional studies before clinical application.Keywords: Colorectal cancer, miR-32, PTEN, Plasma biomarker, Early detection Introduction Colorectal cancer (CRC) is the third most commonly diagnosed cancer and the second leading cause of cancer-related death worldwide, accounting for 10.2% of all cases and 9.2% of all cancer deaths across both sexes [1]. In Iran, CRC ranks as the third most prevalent malignancy, with an age-standardized incidence rate of 15.1 per 100,000 person-years [2, 3]. Although the five-year survival rate for CRC has notably improved and there have been significant advancements in diagnosis and treatment, it remains a significant concern as a leading cause of cancer-related mortality and a major public health challenge [4]. Early detection of CRC results in more effective treatment, increased survival rates, lower medical costs, and improved quality of life for patients [5]. Epigenetic biomarkers, including microRNAs (miRNAs), have emerged as promising tools for malignancies and predicting disease progression. Changes in miRNA levels and function have been detected in numerous tumor types [6]. These short non-coding RNAs (19–25 nucleotides) function as post-transcriptional regulators and can act either as oncogenes or tumor suppressors [7, 8]. MiR-32 (Accession Number MI0000090) is an intronic miRNA located on chromosome 9 [9]. Its expression pattern varies across malignancies: it is upregulated in prostate [10] and gastric [11], while downregulated in other tumor types, including Breast [12], Ovarian [13], and Lung cancers [14]. Phosphatase and tensin homolog (PTEN), a dual-specificity lipid and protein phosphatase, is a critical tumor suppressor and is commonly mutated, deleted, or epigenetically silenced in numerous types of human tumors. PTEN controls key biological processes, including cell survival, proliferation, senescence, angiogenesis, apoptosis, and metabolism [1, 15]. Loss or dysregulation of PTEN results in persistent PI3K/Akt cascade activation, which represents an oncogenic pathway associated with CRC [16]. In this study, we examined the expression levels of miR‑32 and its target gene PTEN in both CRC tissues and plasma. We assessed the diagnostic performance of plasma miR‑32 by comparing its sensitivity and specificity with CEA and CA19‑9, and analyzed its association with clinicopathological features. Materials and methods Patients and clinical samples Fifty-five paired CRC tissues and adjacent normal tissues, along with plasma samples from 55 CRC patients (25 males, 30 females; mean age 57.92 ± 17 years) and 55 healthy controls (27 males, 28 females; mean age 57.94 ± 12 years), were collected at Imam Reza Hospital, Kermanshah, Iran. Samples were immediately processed and stored at − 80 °C until analysis. None of the patients had received chemotherapy, radiotherapy, or adjuvant therapy before surgery. Tissue samples Samples of tumor tissues and matched adjacent normal tissues were collected from CRC patients during surgical resection. Normal tissues were excised from sites located at least 10 cm away from the tumor margin. All samples were rapidly frozen in liquid nitrogen and stored at − 80 °C until analysis. Plasma samples Peripheral blood samples from CRC patients and healthy controls were collected in EDTA-K2 anticoagulant tubes. Following collection, samples were centrifuged to separate plasma, which was aliquoted into two sterile tubes, rapidly frozen in liquid nitrogen, and stored at − 80 °C until analysis. RNA Extraction and cDNA Synthesis Total RNA, including miRNA, was extracted from tissue and plasma samples using the NucleoSpin® miRNA Plasma MN (Macherey-Nagel) RNA extraction kit and the favorgen mRNA kit (Cat. No: FAMIK002) according to the manufacturer’s instructions, respectively. The purity and concentration of RNA were evaluated using a NanoDrop 1000 (Thermo Fisher Scientific, USA), and only RNA samples with A260/A280 ratios between 1.8 and 2.1 were used for subsequent analyses. Polyadenylation of RNA followed by immediate cDNA synthesis was performed using the BON miR Kit (Bon Yakhteh, Tehran, Iran) according to the manufacturer’s protocol. Quantitative real-time PCR Primers specific for miR-32 and the internal control U6 were synthesized by Bonyakhteh Company (Tehran, Iran) and purified to HPLC grade. Expression levels of miR‑32, PTEN, and reference gene U6 were quantified using the BON miR SYBR® Green protocol (Bon Yakhteh, Tehran, Iran; catalog number BON209002) on an Applied Biosystems 7500 Real-Time PCR System (Foster City, CA, USA). The following primer sequences were utilized in this study: U6-forward:5’-CTCGCTTCGGCAGCACATA-3’, U6-Reverse: 5’-CGAATTTGCGTGTCATCCT-3’, PTEN-forward:5’-TTTGAAGACCATAACCCACCAC, PTEN-Reverse:5’-ATTACACCAGTTCGTCCCTTTC-3’. Relative expression levels were determined using the 2−ΔΔCt method. For tissue samples, miR-32 expression was normalized to U6 small nuclear RNA. For plasma samples, U6 was used as an endogenous reference based on prior reports; however, its stability in circulating biofluids may be variable. Future studies should consider validated exogenous spike-in controls or multiple reference miRNAs for normalization. Biomarker detection Plasma levels of CEA and CA19-9 in CRC patients and healthy controls were measured using the Liaison® chemiluminescence analyzer (Diasorin, Saluggia, Italy) following the manufacturer’s protocol provided with the Liaison CEA kit (Cat. No. 314311, Diasorin). Statistical analysis Real-time PCR data were processed using ABI 7500 SDS software v1.3.1 (Applied Biosystems, USA). The Relative Expression Software Tool (REST 2009) was applied to calculate the relative expression of miR-32 between CRC patients and healthy controls. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic performance of miR-32, CEA, and CA19-9. Formal statistical comparisons between AUCs were not performed in the present study. All statistical analyses were conducted using SPSS software v16.0, and a P-value < 0.05 was considered statistically significant. Results Upregulation of miR-32 in colorectal cancer tissues Quantitative real-time PCR analysis was conducted on 55 paired colorectal cancer tissues and their corresponding adjacent non-tumorous tissues to assess miR-32 expression levels. REST analysis demonstrated a significant upregulation of miR-32 in CRC tumor tissues compared with matched adjacent normal tissues (relative expression = 32.0; P < 0.001; Fig. 1A). Differential expression of miR-32 in colorectal cancer. A ΔCt values of miR-32 in CRC tumor tissues compared with matched adjacent normal tissues. B ΔCt values of circulating miR-32 in plasma samples from CRC patients and healthy controls. Lower ΔCt values indicate higher relative expression. Differences between groups were statistically significant (P < 0.001) Upregulation of circulating miR-32 in plasma samples of CRC patients REST analysis demonstrated a significant upregulation of circulating miR-32 in CRC patients compared with healthy controls (relative expression = 89.6; P < 0.001; Fig. 1B). Expression of miR-32 and clinicopathological parameters Analysis of the relationship between miR-32 expression levels and clinicopathological parameters revealed that no statistically significant associations were observed between miR-32 expression and age, sex, tumor size, tumor location, or tumor type (P > 0.05). However, a significant correlation was observed between elevated miR-32 levels and advanced clinical stages of CRC (Stage III/IV vs. Stage I/II, P = 0.026, Table 1). These findings suggest that miR-32 expression may be linked to CRC progression. Table 1Association between miR-32 expression levels and clinicopathological characteristics of CRC patients in tissue and plasma samplesThe expression of miR-32 in tissueThe expression of miR-32 in plasmaVariableNumberMean ± SD*P valueMean ± SDP valueAge(years) 15 50 ≤7.16 ± 1.490.2818. 18 ± 1.850.632 >50406.64 ± 1.277.68 ± 2.64Sex Male256.79 ± 1.310.4937.31 ± 2.180.985 Female306.88 ± 1.747.56 ± 2.98Tumor type Adenocarcinoma327.11 ± 1.530.3588.03 ± 1.730.503 Mucinous236.63 ± 1.127.53 ± 3.31Clinical stages Stage I, II377.29 ± 1.400.0267.25 ± 2.590.326 Stage III, IV186.18 ± 1.168.25 ± 1.48Tumor size ≤5 cm396.58 ± 1.270.1357.84 ± 3.000.883 >5 cm167.28 ± 1.487.95 ± 1.56Tumor location Distal197.16 ± 1.740.5777.33 ± 2.910.414 Proximal366.68 ± 1.278.06 ± 2.15*Standard deviation Association between miR-32 expression levels and clinicopathological characteristics of CRC patients in tissue and plasma samples *Standard deviation Diagnostic potential of miR-32 expression as a biomarker To assess the diagnostic performance of miR-32 expression levels in distinguishing colorectal cancer (CRC) patients from healthy controls, ROC curve analysis was conducted for both tissue and plasma samples. In tissue specimens, miR-32 yielded an area under the curve (AUC) of 0.942 (95% CI, P < 0.001), with a sensitivity of 69% and a specificity of 98%. Similarly, plasma samples demonstrated an AUC of 0.846 (95% CI, P < 0.001), with sensitivity and specificity values of 66% and 98%, respectively (Fig. 2). These findings indicate that miR-32 expression may serve as a promising candidate for the detection of CRC, particularly due to its high specificity in both sample types. Receiver operating characteristic (ROC) curves for miR-32 in colorectal cancer. A ROC curve of miR-32 expression in CRC tissue samples (AUC = 0.942). B ROC curves for plasma biomarkers in colorectal cancer, comparing circulating miR-32 (AUC = 0.846), CEA (AUC = 0.808), and CA19-9 (AUC = 0.857) for distinguishing CRC patients from healthy controls Downregulation of PTEN expression in colorectal cancer tissues To investigate the potential regulatory association between miR-32 and its predicted target gene, PTEN, we quantified PTEN expression levels in paired CRC and adjacent non-tumorous tissues. The analysis revealed a statistically significant reduction in PTEN expression in tumor samples, showing a 3.94-fold decrease compared to normal tissues (P < 0.001; 95% CI: 1.9–7.8), consistent with its downregulation in colorectal malignancies. Evaluation of miR-32 compared to conventional CRC biomarkers ROC curve analysis demonstrated that miR-32 effectively discriminated CRC patients from healthy controls in both tissue and plasma samples. In tissue specimens, miR-32 exhibited strong diagnostic performance (AUC = 0.942). In plasma samples, miR-32 showed good diagnostic accuracy (AUC = 0.846), which was comparable to CA19-9 (AUC = 0.857) and slightly higher than CEA (AUC = 0.808) (Fig. 2). At the optimal cutoff value, the sensitivity of miR-32 was 66%, compared with 46.8% for CEA and 43.2% for CA19-9. The corresponding specificity values were 95% for miR-32, 100% for CEA, and 56.8% for CA19-9. Although plasma miR-32 demonstrated numerically higher sensitivity than conventional markers, direct statistical comparison between AUCs was not performed in the present study. Evaluation miR-32 as a liquid biopsy for colorectal cancer To determine whether miR-32 can be used as a liquid biopsy for colorectal cancer, ROC curve analysis was conducted plasma samples and sensitivity and specificity were calculated for stage I/II and stage III/IV. In stage I/II, the sensitivity was 66.7%, the specificity was 97.3%, and the AUC was 0.776. In stage III/IV, the sensitivity was 62.5%, the specificity was 100%, and the AUC was 0.827 (Fig. 3). ROC curves for Stages I/II and Stages III/ IV in CRC plasma samples. A ROC curve of Stages I/II (AUC = 0.776). B ROC curve of Stages III/ IV (AUC = 0.827) These findings indicate that diagnostic value of miR-32 in plasma does not differ significantly in different stages of the disease. This may be due to the small size of samples, especially in stage I/II. Therefore, further studies with larger sample sizes are required to investigate the diagnostic potential of miR-32 as a biomarker for colorectal cancer in plasma samples. Discussion Colorectal cancer is often diagnosed at an advanced stage, as its symptoms are typically nonspecific, delaying detection and reducing the chances of curative treatment [17]. Among epigenetic mechanisms, microRNAs (miRNAs) play key post-transcriptional roles in regulating gene expression during colorectal tumorigenesis [18]. Their small size and stable hairpin-loop structure make them resistant to degradation, allowing detection in tissue, blood, urine, and stool, and making them promising non-invasive biomarkers [19]. miRNAs are estimated to regulate more than 90% of human genes [14], and their dysregulation contributes to key oncogenic processes, including apoptosis, differentiation, proliferation, growth, migration, invasion, and metastasis [20, 21]. In the present study, miR-32 was significantly upregulated in both CRC tissues and plasma, whereas PTEN, a well-established tumor suppressor, was downregulated in tumor tissues. These findings are consistent with previous reports indicating that elevated miR-32 promotes proliferation, migration, invasion, and resistance to apoptosis by suppressing PTEN through post-transcriptional regulation [22, 23]. PTEN is frequently inactivated in various malignancies, including breast, gastric, bladder, and prostate cancers, leading to constitutive activation of the PI3K/AKT signaling pathway and subsequent tumor progression [23, 24]. Beyond PTEN, miR-32 also targets additional tumor suppressors in CRC. BMP4, TOB1, and OTUD have each been validated as downstream targets, supporting the oncogenic potential of miR-32 through multiple regulatory pathways [7, 25, 26]. However, opposing findings have been reported in some CRC cell lines, where miR-32 appeared to function as a tumor suppressor by downregulating E2F5 [27]. Similarly, Ye et al. demonstrated that miR-32-5p was downregulated in CRC and negatively correlated with SNHG14, suggesting that SNHG14 acts as a molecular sponge to inhibit miR-32-5p activity [28]. These discrepancies may reflect differences in methodology, sample type (cell lines vs. tissues), tumor stage, subtype, or microenvironmental factors, highlighting the complexity of miR-32 regulation. Importantly, our results revealed a stage-dependent increase in miR-32 expression, supporting its potential as a prognostic biomarker in CRC. Although no significant relationships were detected with age, sex, tumor size, or location, the association with advanced tumor stages suggests a possible role for miR-32 in CRC progression and clinical stratification. Although an inverse association between miR-32 and PTEN expression was observed, this study does not provide direct mechanistic evidence of a regulatory interaction. Functional assays, such as luciferase reporter experiments and gain or loss of function studies, are required to confirm whether PTEN is a direct target of miR-32 in CRC cells. Therefore, our findings should be interpreted as correlative rather than causal. The observed stage-dependent increase in miR-32 suggests a potential association with disease progression. However, the relatively small sample size and single center design limit the generalizability of these findings, highlighting the need for larger, multicenter validation studies. Collectively, our results support an oncogenic role for miR-32 in CRC, primarily through PTEN suppression and possibly other tumor suppressor pathways. Further studies are warranted to clarify the molecular mechanisms underlying its context-dependent functions and to evaluate its clinical utility as a circulating biomarker. Conclusions Our findings indicate that miR-32 is upregulated in CRC tissues and plasma and exhibits an inverse association with PTEN expression. These results support a potential oncogenic role for miR-32 in CRC and suggest that circulating miR-32 may represent a promising non-invasive biomarker candidate. Nevertheless, clinical applicability requires validation in independent, multicenter cohorts and functional confirmation of mechanistic interactions. Limitations This study has several limitations. First, the relatively small sample size and recruitment from a single center may limit the generalizability of the findings. Larger, multicenter studies across diverse populations are warranted to validate the diagnostic and prognostic utility of circulating miR-32 and its association with PTEN in CRC. Second, the absence of functional validation experiments, such as luciferase reporter assays, precludes causal interpretation of the observed miR-32–PTEN association. Third, the use of U6 as an internal reference for plasma miRNA quantification represents a methodological limitation, as its stability in circulating samples remains debated. In addition, the confidence intervals generated by REST analysis were wide, likely reflecting inter-individual variability and the modest sample size. To avoid misinterpretation, we report REST-derived relative expression values and P-values and interpret these findings cautiously. Finally, direct statistical comparison of AUCs between miR-32 and conventional biomarkers (CEA and CA19-9) was not performed, which represents a further limitation. Prospective, multicenter studies incorporating optimized normalization strategies and mechanistic assays are required to confirm clinical utility. Acknowledgements The authors gratefully acknowledge the support and assistance provided by the staff of the Medical Genetics Laboratory, the Reference Laboratory, and Kermanshah University of Medical Sciences during the course of this research. Abbreviations Colorectal cancer MicroRNA 32 Phosphatase and tensin homolog Carcinoembryonic antigen Receiver operating characteristic Real-time polymerase chain reaction Authors’ contributions Bayan Mehrjoei and Nayebali Rezvani designed the research. Homayoon Bashiri performed tissues sampling; Nayebali Rezvani and Susan Tahmasebi analyzed data; Bayan Mehrjoei and Parmida Rezvani data collection and wrote the paper; Lida Haghnazari edited and revised the manuscript. Funding This work was financially supported by the Kermanshah University of Medical Sciences, Kermanshah, Iran (Grant #980454). Data availability All data generated or analyzed during this study are fully presented within this published article. Declarations Ethical approval and consent to participate The study protocol was approved by the Ethics Committee of Kermanshah University of Medical Sciences (IR.KUMS.REC.1398.308) and conducted in compliance with the Declaration of Helsinki; written informed consent was obtained from all participants. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All data generated or analyzed during this study are fully presented within this published article.