CEA-delta could be a biomarker of tumor phenotype, clinical stage, and chemotherapeutic response in rectal cancer with OCT4-positive cancer stem cells

Background There is very limited evidence on biomarkers for evaluating the clinical behavior and therapeutic response in rectal cancer (RC) with positive expression of cancer stem cells (CSCs). Methods An exploratory prospective study was conducted, which included fresh samples of tumor tissue from 109 patients diagnosed with primary RC. Sociodemographic, pathological and clinical characteristics were collected from medical records and survey. The OCT4 protein was isolated using the Western Blot technique. It was calculated the ΔCEA, ΔOCT4, and ΔOCT4/GUSB values by assessing the changes before and after chemotherapy, aiming to evaluate the therapeutic response. Results Patients had an average age of 69.9 years, with 55% (n=60) being male. Approximately 63.3% of the tumors were undifferentiated, and the most frequent staging classification was pathological stage III (n=64; 58.7%). Initial positive expression was observed in 77.1% of the patients (n=84), and the median ΔCEA was -1.03 (-3.82 - 0.84) ng/ml, with elevated levels (< -0.94 ng/ml) found in 51.4% of the subjects (n=56). Being OCT4 positive and having an elevated ΔCEA value were significantly associated with undifferentiated tumor phenotype (p=0.002), advanced tumor progression stage (p <0.001), and negative values of ΔOCT4 (p <0.001) (suggestive of poor therapeutic response) compared to those without this status. Conclusion This study identified a significant and directly proportional association among the values of ΔCEA, ΔOCT4, and ΔOCT4/GUSB. These findings suggest that ΔCEA holds potential as a clinical biomarker for determining the undifferentiated tumor phenotype, advanced clinical stage, and poor therapeutic response in RC with CSCs positive expression.


Introduction
Colorectal cancer (CRC) is one of the cancers associated with the highest burden of disease worldwide, with approximately 2 million new cases diagnosed and nearly 1 million deaths reported annually (1)(2)(3).It is considered the third most common malignancy and the second most deadly on a global scale.According to expert projections, it is estimated that by the year 2040, both the incidence and mortality rates of CRC will double, posing a significant threat to healthcare systems worldwide (2).The behavior of this disease varies, being more aggressive in regions where the Western diet is prevalent and disparities in healthcare access exist (4).Consequently, various international organizations emphasize the importance of addressing CRC as a priority in global public health and emphasize the need for research to aid in the timely identification and treatment of this malignancy (5)(6)(7).This research should utilize innovative resources like translational research (8) and precision medicine (9).
Rectal cancer (RC) specifically represents a more aggressive subtype of CRC, characterized by rapid progression and an unfavorable phenotype compared to other colon locations (10)(11)(12).Genetic intratumour heterogeneity (gITH) underlies this phenomenon, resulting from the interplay between genetic and epigenetic variations (tumor biology) and transcriptional plasticity (13).This culminates in population-specific phenotypic plasticity (14).Consequently, extrapolating the effects of chemotherapy treatments for RC management across populations with distinct genetic and epigenetic characteristics poses significant challenges, explaining the variations in response rates and treatment success (13,14).Therefore, understanding the tumor biology of each population becomes imperative to implement precision medicine and identify valuable biomarkers.Cancer stem cells (CSCs) act as a foundation for tumor growth, metastasis, and therapy resistance (15).They represent an intriguing target in translational cancer research, enabling personalized treatment for public health priority cancers.Nevertheless, the variation in gene expression of CSCs in RC remains poorly studied.
Octamer-binding transcription factor 4 (OCT4) is a critical protein involved in the self-renewal of stem cells and is expressed during embryonic phases.In adults, OCT4 has been associated with numerous malignancies as a marker of CSCs expression (16).Tumors expressing OCT4 may exhibit a more aggressive behavior, with rapid evolution and adaptability to the tumor microenvironment, resulting in a poor chemotherapeutic response.Some previous studies have mainly explored OCT4's clinical significance in terms of clinical staging, progression, and chemotherapeutic response (17)(18)(19).However, these studies primarily focus on CRC rather than RC.Moreover, biomarkers capable of identifying OCT4 expression in tumors, which could be valuable in clinical practice, have not yet been investigated.
The oncofetal antigen known as carcinoembryonic antigen (CEA) is expressed in colorectal epithelial cells and serves the purpose of intercellular adhesion and cell recognition (20).Its origin lies in the embryonic endodermal epithelium of the fetus, and its expression is regulated by oncofetal genes (oncofetal antigens are proteins which are typically present only during fetal development but are found in adults with certain kinds of cancer) (21)(22)(23)(24)(25)(26)(27), among which OCT4 could be linked.Following birth, traces of CEA persist in colorectal tissue, yet its levels in the bloodstream are minimal.CEA has been identified as a prognostic biomarker for CRC due to its increased expression and secretion into the blood during invasion and tissue damage (28).Nevertheless, the relationship between CEA and oncofetal genes in terms of affinity and sensitivity remains unclear.Establishing the association between CEA and OCT4 would provide valuable evidence for the identification of gITH and phenotypic plasticity, facilitating personalized approaches and chemotherapy treatments for RC based on clinical progression and chemotherapeutic response.Therefore, considering the global significance of identifying dependable and reproducible biomarkers for RC diagnosis and prognosis, as well as supporting the development of precision medicine treatments, this study aimed to explore the association between serum CEA behavior, tumor phenotype, clinical evolution and chemotherapeutic response in individuals with RC while considering OCT4 status.

Patient selection and sample collection
An exploratory prospective study was conducted, which included fresh samples of tumor tissue and from 109 patients diagnosed with primary RC who were diagnosed at two tertiary referral centers in the city of Cartagena, Colombia, from 2018 to 2023.
Patients included in the study met the following criteria: 1) They agreed to participate by providing informed consent and donating the sample during the diagnostic colonoscopy or followup (before any chemotherapy treatment for the first sample); 2) Their complete medical history was available to verify the data obtained during the interview; 3) They attended evaluation, followup, and treatment by oncology and oncological surgery departments; 4) They had undergone at least 2 sessions of chemotherapy; 5) The medical records contained documented CEA values before and after the chemotherapy sessions; 6) They have not undergone curative or palliative surgery.Patients with a different diagnosis of primary gastrointestinal cancer, who have previously received chemotherapy and those who reported any of the following conditions around the time of blood sample collection for CEA analysis were excluded: 1) Surgical or non-surgical acute gastrointestinal disease (diverticulitis, inflammatory bowel disease, ulcer, pancreatitis); 2) Liver diseases; and 3) Chronic obstructive pulmonary disease.All samples met the appropriate size criteria for molecular analysis.Prior to any treatment, the first sample was obtained during the diagnostic colonoscopy, and the second sample was acquired during the follow-up colonoscopy subsequent to the chemotherapy sessions.The collected fresh tissues were embedded in RNAlater ™ and stored at -80°C for subsequent analysis.

Data collection and variables
Sociodemographic, pathological and clinical characteristics were collected from medical records and structured survey.In this study, the tumor tissues were evaluated by expert pathologists using hematoxylin and eosin slides and documented in the pathological analysis reports.
Data such as age, sex, main symptom and clinical history, tumor differentiation grade, lymph node involvement, presence of metastasis, and TNM stage were collected.In the study, tumors with characteristics described as moderately or poorly differentiated were considered undifferentiated.The description of clinicalpathological staging followed the subcategories of stages (Tis [0], I, IIA, IIB, IIC, IIIA, IIIB, IIIC, IVA, IVB, IVC) based on the eighth edition of the American Joint Committee on Cancer (AJCC) staging system (29).Pathological staging, however, encompassed only the four major groups (I, II, III, IV).Clinical staging was determined by the stage system and the histopathological report, categorizing tumors as local/early, regional/advanced, or metastatic/advanced.Hence, the progression status was defined as early or advanced (regional or metastatic).CEA levels were obtained from the medical records (The CEA value after chemotherapy was derived from the average of all CEA values within 6 months after the completion of at least 2 chemotherapy sessions), and the delta was calculated as follows: CEA before chemotherapy -CEA after chemotherapy = DCEA.In accordance with previous studies that have established cut-off scores (30,31), a reference value of 0.94 was utilized (elevated DCEA for < -0.94 and normal DCEA for > -0.94).In this case, the elevated DCEA value suggests progression and poor therapeutic response following chemotherapy.In addition, DOCT4 and DOCT4/GUSB were calculated to evaluate the response in cancer stem cell marker expression after chemotherapy.GUSB was used as a control for the validity of the molecular techniques (western blot).These calculations followed the same equation as DCEA (pre-chemotherapy value -post-chemotherapy value).Thus, a negative DOCT4 and DOCT4/GUSB values indicate increased expression following chemotherapy, which will be interpreted as a poor therapeutic response.

Western blot analysis
The OCT4 protein was isolated using the Western Blot technique.The tissues were thawed and resuspended in lysis buffer [20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 10% glycerol, 0.2% Nonidet P-40, 1 mM EDTA, 1 mM EGTA, 1 mM phenylmethylsulfonyl fluoride (Sigma-Aldrich, St. Louis, MO, USA), 10 mM NaF, 5 mg/ml aprotinin (Sig-ma-Aldrich), 20 mM leupeptin (Sigma-Aldrich, St. Louis, MO, USA), and 1 mM sodium orthovanadate (Sigma-Aldrich, St. Louis, MO, USA)].The concentration of total protein in the supernatant was quantified using a spectrophotometric method (32).The absorbance at 595 nm was calculated using a standard curve previously prepared with bovine serum albumin (BSA).The samples were prepared with Laemmli loading buffer and denatured by heating at 95°C for 5 minutes.30 mg of protein were loaded onto a 10% polyacrylamide gel prepared with sodium dodecyl sulfate (SDS-PAGE) and subjected to electrophoresis in the presence of an electrophoresis buffer at a constant voltage (33).After electrophoresis, the proteins were transferred from the gel to a PVDF membrane (iBlot ™ Transfer Stack, PVDF Invitrogen ™ Thermo Waltham, MA, USA) using the iBlot ™ 2 Gel Transfer Device dry transfer technology (Thermo Scientific ™ , Waltham, MA, USA).Ponceau staining was performed to confirm that the transfer was successful.Subsequently, the membrane was treated with a blocking solution of 5% skim milk in TTBS1X [10 mM Tris/HCl, 150 mM NaCl, 0.05% Tween-20 (pH 7.5)], and then incubated overnight with the primary antibody anti-OCT4 diluted 1:2000 (Abcam, Cambridge, UK) [EPR2054] (ab109183).After the incubation period, the membrane was washed with TTBS 1X to remove excess antibody.The membrane was then incubated with a horse-radish peroxidaseconjugated secondary antibody for 2 hours.Finally, the membrane was washed with TTBS 1X and the immunodetection was performed using the Super-Signal ™ West Pico chemiluminescent substrate (Thermo Scientific ™ , Waltham, MA, USA).The results were validated using anti-beta glucuronidase antibody 1:2000 (Abcam Cambridge, UK) [EPR10616] (ab166904) as a housekeeping antibody.The PVDF mem-branes were analyzed using an imaging documentation system, using the iBright CL1000 equipment (Thermo Scientific ™ Waltham, MA, USA).The iBright analysis software desktop version (Thermo Scientific ™ Waltham, MA, USA) was used to measure the band densitometry to determine OCT4 expression.

Statistical analysis
The normality of quantitative variables was tested using the Kolmogorov-Smirnov test.Data were presented as mean ± standard deviation (SD) for continuous variables and median (interquartile, IQR) for skewed variables.Qualitative variables were summarized using frequency and percentages.Comparative analysis was carried out using Pearson's Chi-square test or Fisher's exact test for categorical variables and Mann-Whitney or Kruskal-Wallis test for quantitative variables.To identify the variables associated with DCEA, an unadjusted logistic regression was conducted.Furthermore, an exploratory analysis was carried out using linear regression to ascertain any potential association between the quantitative values of DOCT4 and DOCT4/GUSB with DCEA.In addition, the calculation of the prevalence ratio (PR) was performed to evaluate the strength of the association between categorical variables and the outcome of DCEA across different subgroups.Pearson's or spearman's correlation coefficient tests were used for the evaluation of potential correlations between the DOCT4 and DOCT4/GUSB level of expressions and DCEA.A pvalue <0.05 was considered statistically significant.All analyses were performed using Statistical Package for the Social Sciences (SPSS) version 29.0 software.

Ethical statements
This study was approved by the Ethics Committee of the Universidad de Cartagena (Minutes No. 108, 10 May 2018), and was conducted in accordance with the principles of the Helsinki Declaration.Each eligible participant signed an informed consent form.

Patients' characteristics
The patients had an average age of 69.9 years, with 55% (n=60) being male and primarily originating from urban areas (n=94; 86.2%).Gastrointestinal bleeding (n=74; 67.9%) was the main clinical manifestation, followed by intestinal obstruction and acute abdominal pain (n=10; 9.2% for both cases).At the time of diagnosis, the majority of subjects had a normal weight (n=84; 77.1%).The most common personal history was smoking (n=30; 27.5%), followed by alcoholism (n=20; 18.3%), and only 10 individuals had a family history of CRC (Table 1).

Elevated
The presence of an elevated DCEA in conjunction with OCT4 positivity was found to exhibit a significant association with negative values of DOCT4 and DOCT4/GUSB (p <0.001 for both cases) when compared to OCT4 negativity, suggesting a potential link to a suboptimal therapeutic response (Table 8).Lastly, upon evaluating the linear regression analysis examining the relationship between quantitative OCT4 expression values and the deltas of OCT4 and OCT4/GUSB with DCEA, it was determined that while all three variables showed a statistically significant association with predicting the outcome (p <0.001 in all cases), the highest R 2 value attained was only 0.42 (in the case of DOCT4).Furthermore, this association was also observed in the correlation analysis of the quantitative values between DCEA and OCT4 (Rho -0.45; p <0.001), DOCT4 (Rho 0.65; p <0.001), and DOCT4/GUSB (Rho 0.61; p <0.001).

Discussion
The comprehension of the tumor microenvironment, gITH (13), and genotype-phenotype correlation (11) in cancer research establishes a fundamental basis for the identification of targets that permit precise, safe, and effective inhibition of tumorigenesis, migration, differentiation, metastasis, and therapeutic resistance in cancer (10,11,13,14).CSCs represent a crucial element in the tumor microenvironment, demonstrating an unimaginable adaptive capacity against external agents that favors the development of aggressive phenotypes, rapid progression, and a poor therapeutic response (34,35).While numerous studies have identified potential molecular biomarkers in diverse cancers, the genetic diversity of populations and the phenotype resulting from the interaction between the environment and genotype/epigenotype hinder the potential extrapolation of these molecular biomarkers (36).In particular, limited evidence exists in RC regarding molecular biomarkers indicating CSC expression, which could be utilized for personalized chemotherapy treatment in individuals with a more aggressive tumor biology and phenotype.The universal expression of OCT4 in CSCs (as a biological regulator of these cells) has been acknowledged, enabling the differentiation of tumors with pluripotency from those without, ultimately leading to distinct clinical behavior and a worse prognosis (16).
Considering that OCT4 expression could determine the tumor phenotype and therapeutic response in RC (37), how can the evolution and chemotherapy response of RC with CSCs positive expression be evaluated in clinical practice?This assessment can be achieved using a clinically utilized biomarker with quantitative expression dependent on the tumor phenotype and progression.The hypothesis that the quantification of CEA, a glycoprotein whose expression possibly is regulated by oncofetal genes like OCT4, is directly and significantly associated with the quantitative expression of OCT4 in RC has been confirmed by this study.Moreover, the persistent association observed after chemotherapy (represented by the values of DOCT4 and DCEA, indicating the change before and after chemotherapy) provides evidence of its correlation as a clinical biomarker for monitoring and assessing therapeutic response.This finding could help enhance the prognostic performance of RC through the use of combined biomarkers.This is due to the fact that CEA by itself exhibits insufficient performance in predicting the prognosis of RC (sensitivity of 50% and specificity below 80%) (38).
Initially, it was observed that RC exhibited sociodemographic and clinical behavior similar to the literature reports (17)(18)(19)39).It tends to occur more frequently in older adults (around 60-70 years old) probably due to prolonged exposure to risk factors throughout    their lives (40).It also affects individuals from ethnic groups considered vulnerable and at high risk in certain regions (41).For instance, individuals of African descent and Mulattos represented 29 (26.6%) and 35 (32.1%) cases respectively, which can be attributed to socioeconomic inequities, limited healthcare access, and unhealthy lifestyles resulting from difficulties in accessing health education (41).The primary manifestation of RC was gastrointestinal bleeding (n=74; 67.9%), which could be explained

B C A
Red: OCT4 positive expression; Green: OCT4 negative expression.Change in the absolute values of the quantitative parameters: (A) CEA; (B) OCT4; and (C) OCT4/GUSB, before and after chemotherapy, stratified according to the initial expression of the OCT4-cancer stem cell marker.Source: authors.by localized anatomical abnormalities in the rectum.Among the sampled individuals, 77.1% (n=84) had a normal weight range.
Although the reported prevalence of overweight and obesity in the Caribbean region of Colombia is significant (42), this can be explained by unreported weight loss at the time of diagnosis.Furthermore, approximately 30% and 20% reported being smokers and frequent alcohol consumers, respectively.These behaviors constitute potential risk factors for cancer, including RC (43).Interestingly, only 9.2% reported a family history of colorectal cancer, suggesting a lack of significant association between cancers related to familial patterns in this region.While adenocarcinoma (95.4%) was the most frequent histological diagnosis, the undifferentiated phenotype (n=69; 63.3%) predominated, with metastasis occurring in only 10 cases at the time of diagnosis.This suggests a strong influence of genotype-phenotype correlation on the tumorigenesis of the rectal tissue cell line of tumor origin (11).The heterogeneous, distinct, and more aggressive pattern of differentiation observed (possibly due to alterations in regulatory markers like OCT4) may explain this phenomenon (16).However, migration is limited, likely due to the absence or inactivation of a necessary cofactor for this process (16,34).On the other hand, invasion behavior is more frequent, which accounts for the predominance of stage III (n=64; 58.7%) and advanced stage (n=74; 67.9%) at the time of diagnosis.Regarding molecular expression during OCT4 diagnosis, positivity was observed in 77.1% (n=84), which explains the predominant behavior of other described clinical and staging variables, such as the undifferentiated phenotype and advanced stage.Importantly, this prevalence correlates with the percentage of individuals who showed negative values for DOCT4 (71.6%) and DOCT4/GUSB (67.9%), indicating a poor therapeutic response.This implies a direct relationship between the initial identification of CSC-positive RC and the expected therapeutic response.
When we compare our results with those reported in the literature so far, we find that, although fewer than five studies have specifically investigated the relationship between OCT4 and RC (and none to date between OCT4 and CEA) (17)(18)(19)39), similar evidence suggests a comparable clinical and sociodemographic behavior.These studies have revealed a male/female prevalence ratio > 1, with cancer often presenting between the ages of 60-70 years, a predominance of moderate to poor degree of differentiation, and a high frequency of stage III cases at the time of diagnosis.It is noteworthy that all of these studies originate from the Asian continent (China, Japan, and Iran) (17,19,39), and the similarities could be explained by the influence of certain risk factors throughout life in colon cancer and RC.The expression of OCT4 in an alternative form of gastrointestinal cancer has been investigated by several other studies, revealing a comparable trend in relation to the prognostic potential of OCT4 (44).Nevertheless, a correlation with CEA has not been established in these findings, and the study cohort has comprised patients subjected to surgical resection.In our instance, the study population included individuals who had not undergone surgical resection (factor directly impacting the cancer's behavior).
Finally, we found a plausible and expected relationship between the medians of DCEA and the degree of differentiation, staging, initial expression of OCT4, DOCT4, and DOCT4/GUSB, indicating significantly lower DCEA values (i.e., post-chemotherapy CEA value higher than the baseline value).This relationship can be explained by tumor plasticity and cellular phenotype, which may exhibit hyperplasia and lack of differentiation in the original cell line.This leads to increased expression of the CEA glycoprotein in tumors with larger size, greater tissue invasion, and poorer degrees of differentiation.These findings support the self-renewal and pluripotency properties observed in tumors with CSCs positive expression, ultimately resulting in rapid progression that hampers early-stage cancer diagnosis.When grouping the elevated DCEA category status (< -0.94 ng/ml) with OCT4 positive expression, these associations persist compared to not having this status.In other words, we can observe a direct proportionality between CEA expression and OCT4 expression.In cases with a poor therapeutic response, where OCT4 expression continues to increase due to cancer progression, an increase in CEA values is also expected.This leads to negative DCEA values.Therefore, DCEA could serve as a clinical biomarker for monitoring tumor behavior and therapeutic response in RC.Then, it correlates directly with the expression of OCT4, DOCT4, and DOCT4/GUSB.
These results offer various translational applications: 1) Evaluating the initial expression of the OCT4 marker in RC samples would be interesting and valuable to determine CSCs positivity in the tumor and consider the potential persistence of pluripotency and tumor plasticity properties.2) Developing an algorithm based on CSC expression status could establish a personalized approach to assess the impact of curative or palliative surgery, radiotherapy, and chemotherapy according to the specific therapeutic regimens employed.3) Seeking to predict the prognosis and survival of individuals with RC and CSCs positive expression could inform a multidisciplinary approach and shorter intervals between chemotherapy sessions or additional drugs targeting specific signaling pathways related to CSCs.4) Considering the risk of adverse events and quality of life in individuals with a poor expected therapeutic response becomes crucial, as chemotherapy significantly impacts the physical and emotional well-being of the patient.Furthermore, identifying specific pathways in these tumors with OCT4 expression could facilitate the use of targeted biological agents to inhibit frequently expressed signaling pathways involved in differentiation, invasion, and self-renewal.This has the potential to significantly impact the survival of affected individuals.Additionally, the fundamental novelty and pertinence of our research hypothesis are rooted in the absence of any description, up until the present moment, of an alternative serum biomarker that may be amalgamated with CEA to amplify diagnostic performance.Nevertheless, the potential for conjunction arises with a molecular examination, exemplified by the OCT4 status.
We acknowledge the small sample size as a limitation, which may introduce biases when comparing different subgroups based on OCT4 expression and DCEA status.Additionally, the exploratory design did not allow for evaluating changes in clinical and oncological parameters after chemotherapy and their correlation with DCEA, DOCT4, and DOCT4/GUSB values.Furthermore, other post-chemotherapy follow-up parameters that could provide a more precise assessment of cancer progression were not included.However, we want to highlight the following strengths of this study: 1) This study is the first to evaluate OCT4 expression in fresh RC samples, unlike previous studies (17,19,39), which enhances the quality of molecular analysis and result accuracy; 2) It established a correlation between the expression of OCT4 and CEA markers in RC; 3) It analyzed changes in OCT4 and CEA expression before and after chemotherapy, providing novel evidence of a potential routine biomarker for RC monitoring in clinical practice; 4) This study established a new association between two potential signaling pathways linked to the germ cell line of rectal tissue (related to CEA and OCT4 expression in the mutagenic and neoplastic state).These findings lay the groundwork for future studies aiming to conduct cell cultures to describe the signaling pathways associated with differentiation, migration, invasion, self-renewal, and other cellular properties in tumor cells with CSCs positive expression.This can lead to the development of new hypotheses or a more precise understanding of the behavior and tumor microenvironment in RC among individuals with these clinical, oncological, and sociodemographic characteristics.

Conclusions
This study identified a significant and directly proportional association among the values of DCEA, DOCT4, and DOCT4/ GUSB.These findings suggest that DCEA holds potential as a clinical biomarker for determining the undifferentiated tumor phenotype, advanced clinical stage, and poor therapeutic response in RC patients with positive expression of the OCT4 CSCs marker.

1
FIGURE 1 Blue: OCT4 positive expression; Red: OCT4 negative expression.Mean value differences of DCEA based on: (A) Degree of differentiation; (B) Clinicalpathological stage; (C) Pathological stage; (D) Clinical stage; (E) Progression stage; and (F) Therapeutic response (DOCT4 positive value vs. DOCT4 negative value, respectively), stratified according to the initial expression of the OCT4-cancer stem cell marker.Source: authors.

4 3. 36 *
Statistical significance: <0.05. a Statistically significant difference compared to stage I (p=0.003).b Statistically significant difference compared to stage I (p <0.001).c Statistically significant difference compared to stage I (p=0.003).d Statistically significant difference compared to stage I (p=0.004).e Statistically significant difference compared to stage I (p <0.001).f Statistically significant difference compared to local/early stage (p <0.001).g Statistically significant difference compared to local/early stage (p=0.024).
DCEA in rectal cancer with OCT4-positive cancer stem cells

TABLE 1
Baseline characteristics of the study population (N=109).

TABLE 2
Oncological characteristics and molecular expression of the study population (N=109).

TABLE 5
Association between elevated DCEA + OCT4 positive expression status with oncological variables of interest (N=109).

TABLE 6
Association between elevated DCEA + DOCT4 negative value status with oncological variables of interest (N=109).