The expression profiling of serum circPHLPP2 and LncRNA ILF3 in colorectal cancer patients

Abstract

Colorectal cancer (CRC) remains a major cause of cancer-related deaths worldwide, highlighting the need for new non-invasive biomarkers for early diagnosis and prognosis. This study explored the expression of two non-coding RNAs, circular RNA PHLPP2 (circPHLPP2) and long non-coding RNA ILF3-AS1 (lncRNA ILF3-AS1), in CRC patients and assessed their clinical significance. A total of 220 participants were enrolled, including 130 CRC patients (grouped by cancer stage) and 90 healthy individuals. Serum levels of circPHLPP2 and lncRNA ILF3-AS1 were measured using real-time PCR, and traditional tumor markers CEA and CA19-9 were evaluated by immunoassay. Both non-coding RNAs were significantly elevated in CRC patients and showed a progressive increase across cancer stages, reaching the highest levels in metastatic cases. Diagnostic analysis revealed that circPHLPP2 and lncRNA ILF3-AS1 had stronger diagnostic performance than conventional markers. Furthermore, lncRNA ILF3-AS1 was identified as an independent predictor of metastatic CRC. These findings suggest that circPHLPP2 and lncRNA ILF3-AS1 could serve as promising biomarkers for the diagnosis and staging of CRC.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-35356-1.

Introduction

Colorectal cancer (CRC) ranks as the third most common cancer among men and the second among women, responsible for approximately 10% of cancer-related mortality globally¹. In Egypt, It is the third most frequently diagnosed cancer and the second leading cause of cancer-related mortality². With the ongoing evolution of the Middle Eastern economy and shifting lifestyles, the age of CRC diagnosis has progressively decreased. Despite advances in healthcare, a substantial proportion of CRC patients remain asymptomatic in the early stages of the diseases, which often leads to diagnosis at an advanced stage³. This delay in detection is critical, as the 5-year survival rate following resection drops to only 10% in these advanced cases. Thus, early detection and prompt treatment are pivotal to improving the prognosis of CRC patients⁴.

Several factors contribute to the incidence and progression of CRC, including advanced tumor stage, older age, smoking, poor dietary habits, genetic predispositions, and inflammatory bowel diseases⁵.

In recent years, the role of molecular biomarkers in the diagnosis and management of CRC has gained considerable attention⁶.

Among molecular biomarkers, circular RNAs (circRNAs) have emerged as promising candidates due to their stability in blood circulation, resistance to RNase R degradation, and differential expression in various diseases, particularly cancer⁷. Recent studies have highlighted the roles of circRNAs in regulating gene expression and cancer-related processes, including cell proliferation, apoptosis, metastasis, and tumor microenvironment modulation, largely through their interactions with microRNAs and RNA-binding proteins^8,9.

While several circRNAs have been extensively studied in CRC, CircPHLPP2 remains unexplored in this malignancy. Investigating its expression and clinical significance could provide novel insights into CRC pathophysiology and contribute to the development of more effective diagnostic and therapeutic strategies¹⁰.

Long non-coding RNAs (lncRNAs) are transcripts longer than 200 nucleotides that regulate gene expression at transcriptional, post-transcriptional, and epigenetic levels through interactions with DNA, RNA, and protein complexes¹¹. In colorectal cancer, lncRNAs have been implicated in tumorigenesis, metastasis, and chemoresistance via modulation of key signaling pathways, including Wnt, p53, and TGF-β^12,13. Despite these advances, the precise mechanisms underlying the roles of many lncRNAs in CRC remain incompletely understood.

CircPHLPP2 and LncRNA ILF3-AS1 represent promising candidates for CRC research due to their unique roles in cellular stress responses and gene regulation. CircPHLPP2, a recently identified circRNA, has shown potential in regulating key signaling pathways. It contributes to tumor progression and resistance to anti-PD-1 therapy in CRC by decreasing NK cell presence in peripheral blood, limiting their infiltration into tumor tissues, and downregulating granzyme B and IFN-γ expression, ultimately enabling immune evasion¹⁴. Similarly, LncRNA ILF3-AS1 has been implicated in cellular processes such as RNA binding and stability¹⁵, yet its role in cancer remains unexplored.

Recent evidence suggests a functional interplay between circPHLPP2 and ILF3-AS1 in colorectal cancer. CircPHLPP2 has been shown to bind ILF3 protein and regulate IL36γ transcription, thereby promoting tumor growth, immune evasion, and resistance to anti-PD-1 therapy in CRC¹⁴. ILF3-AS1 is a multifunctional RNA-binding factor involved in RNA stability, transcriptional regulation, and stress-responsive signaling pathways, all of which are closely linked to cancer progression¹⁵. Based on this emerging biological connection, we hypothesized that circPHLPP2 and lncRNA ILF3-AS1 may exhibit coordinated expression patterns in colorectal cancer and that their circulating levels could provide complementary clinical information regarding disease presence and progression. Accordingly, the combined evaluation of serum circPHLPP2 and lncRNA ILF3-AS1 was pursued to explore their potential utility as non-invasive biomarkers for CRC.

Given the growing interest in non-coding RNAs as potential biomarkers and therapeutic targets, investigating the expression profiles of both CircPHLPP2 and LncRNA ILF3-AS1 in CRC could provide novel insights into disease pathophysiology and contribute to the development of personalized treatment strategies.

To the best of our knowledge, this is the first study to investigate the expression of circPHLPP2 and lncRNA ILF3-AS1 in serum samples from Egyptian CRC patients. By providing preliminary data on these non-coding RNAs, our findings could serve as a foundation for future studies in different populations and clinical settings.

Subjects and methods

This prospective observational study included 220 participants, divided equally into two groups: 130 colorectal cancer (CRC) patients and 90 healthy controls. CRC patients were recruited from individuals attending the Gastrointestinal Endoscopy Unit at Al-kasr Al-Ainy Hospital, Cairo University, Egypt for colorectal screening.

The sample size was determined through power analysis to ensure 80% statistical power to detect significant differences in RNA expression levels between CRC patients and healthy controls, with an alpha value of 0.05. Participants were randomly assigned to the CRC and control groups using a computer-generated random number sequence to ensure unbiased allocation. Data acquisition and analysis were blinded, with laboratory personnel unaware of the patient group. Additionally, CRC stage scoring was performed by an independent investigator who was blinded to patient identity and group allocation.

Inclusion and exclusion criteria

Patients included in the study had histologically confirmed CRC and had not received prior chemotherapy, radiotherapy, or surgical interventions. Individuals with hereditary CRC syndromes, inflammatory bowel diseases, recurrent CRC, other cancers, precancerous lesions, inflammatory polyps, or intestinal adenomas were excluded. The diagnosis and staging of CRC were determined using colonoscopic findings, imaging studies, and pathological evaluation.

The CRC patients were stratified into three groups based on the American Joint Committee on Cancer (AJCC) staging system¹⁶. The early-stage group (stages I and II) included 45 patients with localized or locally advanced tumors that had not metastasized. The stage III group consisted of 45 patients with regionally advanced tumors involving lymph nodes but without distant metastasis. The stage IV group included 40 patients with distant metastasis.

The control group comprised 90 age- and sex-matched healthy individuals with no history of cancer, major chronic illnesses, or gastrointestinal symptoms. The study was conducted following the principles outlined in the Declaration of Helsinki. Approval was obtained from the Ethics Committee of Al-Kasr Al-Ainy hospital, Cairo University, Egypt with approval number 501-06-24. The privacy rights of human subjects have been observed and written informed consent was secured from all participants prior to enrolment.

Sample collection

A 5 mL sample of venous blood was obtained from each participant using sterile techniques and collected in serum separator tubes. The samples were left to clot at room temperature for 15 min and then centrifuged at 4000 rpm for 10 min. The separated serum was aliquoted and stored at −80 °C until further analysis.

The serum levels of carcinoembryonic antigen (CEA) and carbohydrate antigen 19 − 9 (CA19-9) were quantified using enzyme-linked immunosorbent assay (ELISA) kits following the manufacturers’ protocols. These markers were included to provide complementary diagnostic and prognostic insights and to assess their correlations with the expression patterns of circPHLPP2 and lncRNA ILF3-AS1.

RNA extraction and quantification

Total RNA, including circRNAs and lncRNAs, was extracted from serum using the miRNeasy Mini Kit (Qiagen, Germany) following the manufacturer’s protocol. Serum samples were mixed with QIAzol lysis reagent, followed by phase separation using chloroform. The aqueous phase was mixed with ethanol and applied to silica-based spin columns. RNA was eluted in RNase-free water and quantified using a NanoDrop spectrophotometer. The purity of RNA was assessed by determining A260/280 and A260/230 ratios, with acceptable ranges of 1.8–2.1 and > 1.7, respectively.

Reverse transcription and qRT-PCR

Extracted RNA was reverse-transcribed into complementary DNA (cDNA) using the miScript II RT Kit (Qiagen). The reaction was incubated at 37 °C for 60 min, followed by enzyme inactivation at 95 °C for 5 min. cDNA was stored at −80 °C until use. Quantitative real-time PCR (qRT-PCR) was performed using the miScript SYBR Green PCR Kit and target-specific primers for circPHLPP2, lncRNA ILF3-AS1, and an endogenous control gene. In the present study, the term ‘lncRNA ILF3-AS1’ refers to an antisense long non-coding RNA transcribed from the ILF3 genomic locus. Quantitative real-time PCR primers were designed to specifically target the ILF3-AS1 transcript rather than the protein-coding ILF3 mRNA or its encoded protein.

The reaction was conducted in a Rotor-Gene Q thermocycler (Qiagen) with the following conditions: 95 °C for 15 min for initial activation, followed by 40 cycles of denaturation (94 °C for 15 s), annealing (55 °C for 30 s), and extension (70 °C for 30 s). The relative expression levels of circPHLPP2 and lncRNA ILF3-AS1 were calculated using the ΔΔCt method. Expression fold changes were expressed as relative quantification values, with the control group serving as the reference. A positive fold change indicated upregulation, while a negative value indicated downregulation.

The primer sequences used for quantitative real-time PCR analysis of circPHLPP2, lncRNA ILF3-AS1, and the endogenous control gene GAPDH are provided in Table 1. GAPDH was used as the internal reference gene for normalization of gene expression levels. Mean cycle threshold (Ct) values, amplification efficiencies, and assay linearity (R²) for all qRT-PCR targets are provided in Supplementary Table S1.

Table 1.

Primer sequences used for quantitative real-time PCR analysis.

Target	Primer direction	Sequence (5′−3′)
CircPHLPP2	Forward	TGTATGATTCGATTTTATGGTGAGA
CircPHLPP2	Reverse	GTAGTGGCAGTGGTAGTGTC
LncRNA ILF3-AS1	Forward	TAAACCCCACTGTCTTCC
LncRNA ILF3-AS1	Reverse	TTCCTTGCTCTTCTTGCTC
GAPDH	Forward	TGCACCACCAACTGCTTAGC
GAPDH	Reverse	GGCATGGACTGTGGTCATGAG

Statistical analysis

The data were analyzed using GraphPad Prism software, version 10.4 and The Statistic Package of Social Sciences (SPSS) software, version 24. Results were expressed as mean ± standard deviation (SD) or mean (percentage). The normality of data distribution across groups and subgroups was assessed using the D’Agostino and Pearson omnibus normality test, with p > 0.05 indicating normal distribution. While the overall datasets satisfied normality assumptions, subgroup analyses showed variable distribution patterns; therefore, parametric or non-parametric statistical tests were applied accordingly.

While the overall data for each RNA passed the normality test, variability was observed in subgroup distributions. Depending on the distribution patterns, statistical differences in expression fold changes between groups and subgroups were evaluated using either the student’s t-test or the Mann–Whitney U test, as appropriate.

The correlation coefficient was calculated using Pearson’s product-moment correlation to assess the relationships between the expression levels of circPHLPP2, lncRNA ILF3-AS1, and clinicopathological parameters. A p-value < 0.05 was considered statistically significant. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of the studied RNAs and markers. This included determining diagnostic accuracy, cutoff values, sensitivity, and specificity. The area under the curve (AUC) was used as a measure of diagnostic and prognostic efficiency. Additionally, logistic regression was performed to assess potential predictors of metastatic CRC.

Blinding and randomization

This study followed an observational case-control design without any clinical or therapeutic interventions that require randomization; therefore, no randomization of participants to interventions was performed. To minimize potential bias, laboratory personnel performing RNA extraction and quantitative real-time PCR analyses were blinded to the clinical status of the samples. In addition, statistical analyses were conducted using anonymized datasets in which group allocation was concealed until completion of the primary analyses.

Results

No significant differences were observed in sex distribution and age between groups (Table 2). CEA and CA19-9 levels were significantly elevated in CRC patients compared to controls (P < 0.0001). Notably, CEA levels increased progressively from early stages to stage IV (P < 0.0001 for early stages vs. stage III, P = 0.008 for stage III vs. IV), while CA19-9 showed significant differences between early stages and stage III (P < 0.0001) while no significance was observed between stage III and stage IV (P = 0.16).

Table 2.

Demographic and biochemical characteristics of study participants.

Variable	Control (n = 90)	All CRC patients (n = 130)	Early stages (I and II) (n = 45)	Stage III (n = 45)	Stage IV (n = 40)	P-value
Sex Male Female	42 (46.7%) 48 (53.3%)	60 (46.2%) 70 (53.8%)	18 (40%) 27 (60%)	23 (51.1%) 22 (48.9%)	19 (47.5%) 22 (48.9%)	0.941
Age (years)	60.9 ± 4.54	62.2 ± 4.61	60.7 ± 4.8	62.9 ± 4.83	63.3 ± 4.02*	0.061
CEA (ng/ml)	3.16 ± 1.37	6.29 ± 2.36*	4.3 ± 1.38*	6.79 ± 1.9*¥	7.98 ± 2.1*¥#	0.0001*
CA19-9 (U/ml)	18.9 ± 6.98	28 ± 10.66*	22 ± 7.33*	29.6 ± 6.84*¥	32.9 ± 13.9*¥	0.0001*

CRC: Colorectal cancer; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19 − 9.

Data are expressed as mean (percentage) or mean ± SD.

P-values were calculated by comparing the control group with all CRC patients using either the student’s t-test or the Mann-Whitney U test, as appropriate.

^* Significant from control, ^¥ Significant from early stages (I and II), ^# Significant from stage III.

Serum CircPHLPP2 and LncRNA ILF3-AS1 expression levels were significantly higher in CRC patients compared to healthy controls (Table 3; Fig. 1). CircPHLPP2 expression increased progressively across CRC stages (P = 0.0002 for early vs. stage III, P = 0.01 for stage III vs. IV). LncRNA ILF3-AS1 expression also showed a significant increase from early stages to advanced disease (P < 0.0001 for early vs. stage III, P = 0.0001 for stage III vs. IV).

Table 3.

Serum expression levels of CircPHLPP2 and LncRNA ILF3-AS1.

	Control (n = 90)	All CRC patients (n = 130)	Early stages (I and II) (n = 45)	Stage III (n = 45)	Stage IV (n = 40)	P-value
CircPHLPP2
Mean ± SD (SE)	1.43 ± 0.70 (0.07)	2.97 ± 0.97 * (0.08)	2.36 ± 0.87* (0.13)	3.07 ± 0.77*¥ (0.11)	3.55 ± 0.91*¥# (0.14)	< 0.001*
Median (Min – Max)	1.13 (0.57–3.2)	3.07 (0.75–5.35)	2.39 (0.75–3.95)	3.1 (1.05–4.5)	3.51 (1.78–5.35)
LncRNA ILF3-AS1
Mean ± SD (SE)	1.09 ± 0.39 (0.04)	3.37 ± 1.28* (0.11)	2.33 ± 0.92* (0.14)	3.51 ± 0.91*¥ (0.13)	4.39 ± 1.09 *¥# (0.17)	< 0.001*
Median (Min – Max)	1.02 (0.54–2.71)	3.42 (0.81–6.80)	2.1 (0.81–4.30)	3.6 (1.12–4.95)	4.35 (2.49–6.80)

CRC: Colorectal cancer; SD: Standard deviation, SE: Standard error, Min: Minimum, Max: Maximum.

P-values were calculated by comparing the control group with all CRC patients using either the student’s t-test or the Mann-Whitney U test, as appropriate.

^* Significant from control, ^¥ Significant from early stages (I and II), ^# Significant from stage III.

Fig. 1.

(a) Serum pression of CircPHLPP2 in study groups; (b) Serum Expression of LncRNA ILF3-AS1 in study groups.

Correlation analysis demonstrated a significant positive correlation between CircPHLPP2 and LncRNA ILF3-AS1 expression levels in CRC patients (r = 0.494, P < 0.0001) (Table 4; Fig. 2), suggesting a potential interplay between these biomarkers in CRC progression.

Table 4.

Correlation between CircPHLPP2 and LncRNA ILF3-AS1 expression levels in CRC patients.

Parameter	CircPHLPP2		LncRNA ILF3-AS1
	R	P-value	r	P-value
Age	−0.028	0.745	0.104	0.239
CEA	0.31	0.0003*	0.475	< 0.0001*
CA19-9	0.192	0.029*	0.385	< 0.0001*
CircPHLPP2	-	-	0.494	< 0.0001*
LncRNA ILF3-AS1	0.494	< 0.0001*	-	-

CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19 − 9.

^* : Significant at P < 0.05.

Fig. 2.

Correlation between CircPHLPP2 and LncRNA ILF3-AS1 Expression Levels in CRC Patients.

Receiver operating characteristic (ROC) analysis revealed that CircPHLPP2 and LncRNA ILF3-AS1 exhibited superior diagnostic performance compared to traditional tumor markers CEA and CA19-9 in distinguishing CRC patients from healthy controls (Table 5; Fig. 3): CircPHLPP2: AUC = 0.896 (95% CI: 0.854–0.938), P < 0.0001, LncRNA ILF3-AS1: AUC = 0.967 (95% CI: 0.947–0.988), P < 0.0001, CEA: AUC = 0.882 (95% CI: 0.835–0.929), P < 0.0001 and CA19-9: AUC = 0.771 (95% CI: 0.709–0.834), P < 0.0001.

Table 5.

The ROC analysis for discrimination between CRC cases and controls.

Variables	Cut off	Sn (%)	Sp (%)	AUC	95% CI	P-value
CEA (ng/ml)	> 3.75	86.1	81.1	0.882	0.835 to 0.929	< 0.0001*
CA19-9 (U/ml)	> 18.8	89	58.9	0.771	0.709 to 0.834	< 0.0001*
CircPHLPP2	> 1.97	84.6	81.1	0.896	0.854 to 0.938	< 0.0001*
LncRNA ILF3-AS1	> 1.37	95.4	86.7	0.967	0.947 to 0.988	< 0.0001*
Combined circPHLPP2 + lncRNA ILF3-AS1	-	95.4	85.6	0.973	0.956 to 0.990	< 0.0001*

AUC: Area under the curve; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19 − 9; Sn: Sensitivity; Sp: Specificity; CI: Confidence interval. *: Statistically significant at P < 0.05.

Fig. 3.

Receiver operating characteristic (ROC) curves illustrating the diagnostic performance of (a) CEA, (b) CA19-9, (c) CircPHLPP2, (d) LncRNA ILF3-AS1, and (e) the combined circPHLPP2 and lncRNA ILF3-AS1 model for discriminating CRC patients from healthy controls.

Both CircPHLPP2 and LncRNA ILF3-AS1 exhibited high sensitivity (84.6% and 95.4%) and specificity (81.1% and 86.7%), indicating their potential as diagnostic biomarkers.

To assess the ability of these markers to differentiate metastatic from non-metastatic CRC, ROC analysis was performed (Table 6; Fig. 4).

Table 6.

The ROC analysis for discrimination between metastatic CRC cases and non-metastatic cases.

Variables	Cut off	Sn (%)	Sp (%)	AUC	95% CI	P-value
CEA (ng/ml)	> 6.35	80	65	0.788	0.706 to 0.872	0.001*
CA19-9 (U/ml)	> 27.54	62.5	56.7	0.652	0.538 to 0.766	0.0058*
CircPHLPP2	> 3.17	72.5	54.4	0.722	0.628 to 0.816	< 0.0001*
LncRNA ILF3-AS1	> 3.36	80	62.2	0.814	0.73 to 0.8894	< 0.0001*
Combined circPHLPP2 + lncRNA ILF3-AS1	-	72.5	64.1	0.774	0.702 to 0.845	< 0.0001*

AUC: Area under the curve; CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19 − 9; Sn: Sensitivity; Sp: Specificity; CI: Confidence interval. *: Statistically significant at P < 0.05.

Fig. 4.

Receiver operating characteristic (ROC) curves illustrating the diagnostic performance of (a) CEA, (b) CA19-9, (c) CircPHLPP2, (d) LncRNA ILF3-AS1, and (e) the combined circPHLPP2 and lncRNA ILF3-AS1 model for discriminating metastatic from non-metastatic CRC patients.

LncRNA ILF3-AS1 had the highest AUC (0.814, P < 0.0001), demonstrating better predictive accuracy than CircPHLPP2, CEA, and CA19-9.

CircPHLPP2 also showed a significant AUC of 0.722 (P = 0.0001), followed by CEA (AUC = 0.788, P = 0.001) and CA19-9 (AUC = 0.652, P = 0.0058).

Multivariate logistic regression identified LncRNA ILF3-AS1 (β = 0.362, P = 0.029) and CEA (β = 0.156, P = 0.048) as independent predictors of metastatic CRC (Table 7). In contrast, CircPHLPP2 and CA19-9 were not significant predictors. These findings suggest that LncRNA ILF3-AS1 is the most predictive biomarker for CRC metastasis, followed by CEA. While CircPHLPP2 did not reach statistical significance, its potential role in CRC progression warrants further investigation.

Table 7.

Logistic regression analysis for predicting metastatic CRC.

Variable	β Estimate	Standard Error	95% CI	P-value
Intercept	− 4.75	0.765	−6.312 to −3.306	< 0.0001*
CEA	0.156	0.079	0.002 to 0.312	0.048*
CA19-9	0.011	0.015	−0.018 to 0.039	0.464
CircPHLPP2	0.221	0.188	−0.150 to 0.592	0.239
LncRNA ILF3-AS1	0.362	0.166	0.032 to 0.683	0.029*

CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19 − 9; CI: Confidence interval.

*: Statistically significant at P < 0.05.

Overall, CircPHLPP2 and LncRNA ILF3-AS1 are highly upregulated in CRC patients, correlate with disease progression, and show superior diagnostic performance compared to CEA and CA19-9. LncRNA ILF3-AS1 emerged as the most significant predictor of metastatic CRC, reinforcing its potential as a novel biomarker for CRC diagnosis and prognosis.

Discussion

This is the first study to analyze circPHLPP2 and lncRNA ILF3-AS1 expression in serum from Egyptian CRC patients, providing preliminary data that requires further validation in diverse ethnic populations.

In this study, we investigated the expression patterns of circPHLPP2 and lncRNA ILF3-AS1 in CRC patients and assessed their diagnostic and prognostic potential. Our findings revealed that both circPHLPP2 and lncRNA ILF3-AS1 were significantly upregulated in CRC patients compared to healthy controls. Moreover, their expression levels were progressively higher in advanced disease stages, suggesting their involvement in CRC progression. Notably, both biomarkers demonstrated superior diagnostic performance compared to conventional tumor markers CEA and CA19-9, and lncRNA ILF3-AS1 emerged as an independent predictor of metastatic CRC. These results align with the growing body of evidence supporting the crucial roles of non-coding RNAs in CRC development and progression¹⁷.

The role of circPHLPP2 and lncRNA ILF3-AS1 in CRC remains largely unexplored. However, our findings align with previous research on other non-coding RNAs implicated in CRC. Several studies have reported the upregulation of oncogenic lncRNAs, such as MALAT1, HOTAIR, and H19, in CRC, linking them to tumor progression, metastasis, and chemoresistance¹⁸. MALAT1, for instance, has been shown to promote CRC invasion and migration via modulation of the Wnt/β-catenin pathway¹⁹. Similarly, lncRNA ILF3-AS1 has been identified as a promoter of colon adenocarcinoma progression through the miR-619-5p/CAMK1D axis²⁰, supporting our observation that lncRNA ILF3-AS1 is associated with CRC progression.

Furthermore, ILF3 has been reported to interact with mRNA, leading to an increase in gene expression²¹. Although this function has been described for the ILF3 protein, it supports the biological relevance of the ILF3 genomic locus, from which ILF3-AS1 is transcribed, and may partly explain the dysregulation of ILF3-associated non-coding transcripts in CRC.

Given that lncRNA ILF3-AS1 expression significantly correlated with advanced CRC stages in our study, it is plausible that it may interact with similar oncogenic pathways, warranting further investigation.

The significant correlation between circPHLPP2 and lncRNA ILF3-AS1 expression suggests a potential interplay between these non-coding RNAs in CRC. Although the exact mechanisms remain unclear, circRNAs are known to act as competing endogenous RNAs, sponging miRNAs to modulate gene expression²². In the ceRNA regulatory network, circRNAs compete with mRNA transcripts for miRNA binding. This can protect the mRNA transcripts from miRNA-mediated degradation or translational repression, thereby affecting gene expression²³. It is plausible that circPHLPP2 may interact with specific miRNAs to modulate ILF3-AS1 expression, thereby influencing CRC progression. Several studies have demonstrated circRNA’s ability to influence cancer progression, through different signaling molecules like MAPK/ERK signaling, PI3K/Akt signaling, and Wnt/β-catenin signaling²⁴.

Beyond its epigenetic regulatory functions, emerging evidence suggests that ILF3-associated antisense transcripts may contribute to key oncogenic processes, including epithelial–mesenchymal transition (EMT), immune modulation, and chemoresistance^25,26. Antisense lncRNAs transcribed from cancer-relevant loci have been implicated in EMT through chromatin remodeling and transcriptional repression of tumor suppressor genes, thereby promoting invasive and metastatic phenotypes²⁷. In addition, ILF3-related regulatory networks have been linked to immune-related signaling pathways and cellular stress responses, which may influence tumor–immune interactions within the colorectal cancer microenvironment²⁸. Furthermore, lncRNAs transcribed from the ILF3 locus have been associated with resistance to chemotherapy via epigenetic silencing and modulation of drug-response pathways²⁹.

In this context, the observed correlation between circPHLPP2 and lncRNA ILF3-AS1 expression may reflect complementary aspects of colorectal cancer biology, whereby circPHLPP2 is associated with immune evasion mechanisms, while ILF3-AS1 reflects epigenetic plasticity, EMT activation, and therapeutic resistance³⁰.

Moreover, lncRNAs have been implicated in epigenetic regulation, EMT, and chemoresistance. For example, H19 has been shown to drive EMT and enhance CRC metastasis by acting as a ceRNA for miR-138 and miR-200a³¹. The potential role of ILF3-AS1 in these pathways should be further explored, particularly its involvement in EMT and the regulation of downstream oncogenic targets. LncRNAs also have an important role in the modulation of chemoresistance and are often associated with ATP-binding cassette (ABC) transporters, which are linked to multidrug resistance in CRC³². The interactions of ILF3-AS1 with ABC transporters may also be explored to understand the chemoresistance mechanisms in CRC.

Our findings highlight the potential clinical utility of circPHLPP2 and lncRNA ILF3-AS1 as non-invasive biomarkers for colorectal cancer detection and prognosis. Both biomarkers outperformed CEA and CA19-9 in distinguishing CRC patients from healthy individuals, with lncRNA ILF3-AS1 demonstrating the highest diagnostic performance. In addition, the combined analysis of circPHLPP2 and lncRNA ILF3-AS1 further improved diagnostic discrimination compared to individual biomarkers, supporting the potential value of a dual-marker approach in colorectal cancer detection. These findings suggest that circPHLPP2 and lncRNA ILF3-AS1 may contribute to improved CRC detection and disease monitoring, pending further validation in larger cohorts.

DNA methylation-based assays represent one of the most established liquid biopsy approaches for colorectal cancer screening, with SEPT9 being among the most extensively validated circulating biomarkers. SEPT9 methylation testing has demonstrated clinical utility for CRC detection and has been incorporated into screening strategies in several settings. However, despite their maturity, methylation-based biomarkers may not fully capture tumor dynamics or regulatory RNA networks³³. In this context, circulating non-coding RNAs, such as circPHLPP2 and lncRNA ILF3-AS1, may provide complementary biological information reflecting tumor activity, progression, and metastatic potential rather than serving as direct replacements for existing methylation-based assays.

Furthermore, MALAT1 has been considered a biomarker for metastatic CRC and a potential therapeutic target¹⁹. Similarly, ILF3-AS1 could be further investigated as a therapeutic target, especially given its strong predictive value for CRC metastasis. Previous studies have suggested that targeting oncogenic lncRNAs may provide novel therapeutic strategies for CRC, particularly in patients with advanced or drug-resistant disease³⁴. Studies have shown that targeting lncRNAs by using antisense oligonucleotides in clinical trials has been effective in cancer therapeutics³⁵.

Given these findings, further in vitro and in vivo studies are essential to confirm the causation, sequence of regulatory events, and downstream targets of circPHLPP2 and lncRNA ILF3-AS1 in CRC progression. Functional studies should include knockdown and overexpression experiments to validate their roles in key oncogenic pathways.

Limitations and future directions

While our findings provide valuable insights, this study has several limitations. It is the first to assess circPHLPP2 and lncRNA ILF3-AS1 expression in the serum of Egyptian CRC patients, offering preliminary data that require validation in larger, ethnically diverse populations. Further studies should investigate these biomarkers in different biological samples, such as plasma and tissue, and compare their expression levels in serum and tissue to establish a more comprehensive correlation.

Additional studies are required to clarify the roles of circPHLPP2 and lncRNA ILF3-AS1 in CRC progression. In vitro and in vivo studies are needed to explore their causal relationships, regulatory pathways, and downstream targets. Functional validation through knockdown and overexpression experiments will help determine their involvement in key oncogenic processes. Additionally, mechanistic studies are warranted to examine their interactions with miRNAs and downstream effectors to better understand their regulatory networks. Although tumor staging was performed according to the AJCC system, detailed analyses based on individual TNM subcategories were limited by data availability and could be addressed in future investigations. The present study was cross-sectional in nature, and longitudinal follow-up data were not available; therefore, future prospective studies are needed to evaluate the prognostic value of circPHLPP2 and lncRNA ILF3-AS1 using survival-based analyses. Furthermore, inclusion of patients with colorectal adenomas and other precancerous lesions would be essential to evaluate their utility in early disease detection.

The present study adopted a targeted, hypothesis-driven approach focusing on circPHLPP2 and lncRNA ILF3-AS1 based on emerging evidence of their biological interaction and relevance to colorectal cancer. High-throughput screening approaches, such as RNA-sequencing or microarray analyses, represent an important exploratory step and should be pursued in subsequent investigations to identify additional candidate non-coding RNAs for diagnostic and prognostic validation.

Finally, functional validation experiments should be performed to confirm ILF3-AS1’s regulatory mechanisms in CRC. Further research should also assess whether ILF3-AS1 can serve as a predictor of treatment response or chemoresistance, potentially establishing it as a novel therapeutic target.

Conclusion

In conclusion, our study identifies circPHLPP2 and lncRNA ILF3-AS1 as promising biomarkers for CRC diagnosis and progression. Their significant upregulation in CRC, strong correlation with disease stage, and superior diagnostic performance compared to conventional tumor markers highlight their potential as novel non-invasive biomarkers. Notably, lncRNA ILF3-AS1 demonstrated the highest predictive value for metastatic CRC, making it a prime candidate for further research. Further investigations are necessary to validate these findings, explore their mechanistic roles in CRC, and evaluate their potential as therapeutic targets.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Ahmed Alobaida and Ahmed Elshafei contributed to the conceptualization of the study. Alia D Alshammari performed the formal analysis. Ahmed Alobaida was responsible for funding acquisition. Emad Gamil Khidr conducted the investigation. Ahmed Elshafei and Emad Gamil Khidr performed the methodology. Taslim Alhilal and Alia D Alshammari drafted the manuscript, and Taslim Alhilal and Emad Gamil Khidr contributed to the review and editing process. All authors read and approved the final version of the manuscript.

Funding

This research has been funded by Deputy for Research & Innovation, Ministry of Education through Initiative of Institutional Funding at University of Ha’il – Saudi Arabia through project number (IFP-22 111).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.