Evaluation of the Relationship between CD44 Expression and Gleason Grade among Prostate Adenocarcinoma and Benign Prostatic Hyperplasia: A Cross-Sectional Study

Prostate adenocarcinoma (PAC) is primarily diagnosed through transrectal prostate biopsy [1]. Autopsy studies have shown that over 50% of men between 51 and 60 years of age exhibit pathological features consistent with a benign prostatic hyperplasia (BPH), diagnosis on prostate tissue biopsies [1]. Therefore, accurate differentiation of adenocarcinoma from BPH or normal prostate tissue is clinically crucial. Appropriate decision-making regarding invasive or conservative management of PAC depends on a thorough prognostic assessment [2]. The Gleason scoring system is the primary tool for the pathological evaluation of prostate malignancies, with high prognostic value [2]. It consists of two scores: the first one represents the most predominant histological pattern, and the second score represents the second most predominant pattern in the specimen. A total Gleason score of less than 6 indicates the least clinically significant disease, while a score of more than 8 is associated with the least differentiated tumors and worse prognosis [2]. In 2014, the International Society of Urological Pathology proposed a new classification system based on Gleason scores for improved prognostic evaluation. It consists of five grade groups, each accounting for one or more Gleason scores, which simplifies scoring and improves assessment of prognosis [3]. Despite these advancements, discrepancies between pathologists in the grading of prostatic adenocarcinoma highlight the need to investigate biomarkers that can aid in grading classification and prognosis. Although prostate-specific antigen and Gleason grade groups are the most commonly used biomarkers in practice to evaluate prostate cancer prognosis [4], investigating novel immunohistochemical (IHC) biomarkers is an active area of research in oncological pathology [5‒7]. Tissue-based biomarkers are already available in the market to provide further guidance in risk stratification for confirmed histological diagnoses of prostate cancer, where tissue samples are available for additional testing. These include tests that utilize multiple gene RNA expression levels like Prolaris and tests that use several protein markers like ProMark [8]. Despite these advancements, there remains a need for biomarkers that can more accurately predict disease progression and guide therapeutic decisions [9, 10].

Cluster of differentiation 44 (CD44) is a family of transmembrane glycoproteins that play a crucial role in cell growth, differentiation, adhesion, and migration [11]. Further studies revealed alterations in CD44 conformation and membrane localization can activate a series of downstream signaling cascades that enhance cell proliferation, survival, migration, invasiveness, and chemoresistance [11]. In fact, CD44s (the predominant isoform of CD44) are identified as a metastasis suppressor gene in prostate cancer [12].

Although some studies have examined the relationship of CD44 expression with Gleason score [13, 14], there is a scarcity of studies assessing CD44 expression with Gleason grade groups [15]. The current study, which to the best of our knowledge, is the first in the Iranian patient population that aimed to investigate the correlation of CD44 expression with both PAC Gleason scores and PAC grade groups in a single study with more statistical detail. Furthermore, CD44 expression in PAC is compared to a larger sample of BPH specimens as a control group. Additionally, we also investigated the correlation of CD44 expression with the patient’s age.

This is an analytical cross-sectional study conducted in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. The study was conducted in accordance with the 1964 Helsinki Declaration, and only the medical records and samples of patients who provided their written consent to have their information used for research purposes were utilized. The protocol of this study was reviewed and approved by the Ethics Committee of Shahed University with the approval code “IR.SHAHED.REC.1402.104.”

All prostate tissue specimens from open prostatectomy or transurethral resection of the prostate performed between 2018 and 2023 were retrieved from the pathology database of Mostafa Khomeini Hospital. The patients underwent transurethral resection of the prostate or open prostatectomy for refractory urinary symptoms or complications, such as postrenal acute kidney injury and recurrent urinary retention. Open prostatectomies were also performed for selected patients who were candidates for open surgical management. The age of the patients was also gathered from their records. Specimens were fixed in 10% formalin. Sections with a thickness of 0.4 mm were prepared and stained with hematoxylin and eosin. All the specimens were reviewed by two experienced pathologists to confirm the diagnosis and to assign Gleason score and grade group for PAC specimens. The specimens exhibiting minimal bleeding, necrosis, and a sufficient number of neoplastic cells were chosen for IHC analysis. Those specimens lacking adequate tissue for staining or having incomplete documentation in the hospital pathology database were excluded from this research.

For IHC staining of CD44 in tissue specimens, we used a mouse monoclonal antibody (clone: BC8, isotype: IgG1) manufactured by Biocare Medical, USA. Positive cells were defined based on cytoplasmic/membranous staining. To ensure the accuracy and reliability of the observed staining patterns, IHC controls were utilized. Fallopian tube epithelial cells were used as a positive control because they consistently express CD44, providing a reliable positive control for comparison. This practice is supported by studies that demonstrate the consistent expression of CD44 in fallopian tube epithelial cells [16], making it an ideal positive control. Fallopian tube stromal tissue was used as a negative control. For each specimen, the percentage of cells with positive CD44 expression was recorded separately in the area of the primary Gleason pattern and the area of the secondary Gleason pattern. A weight was assigned to each percentage based on the area occupied by its pertaining pattern in the specimen. The total CD44 expression of the specimen was calculated as the weighted arithmetic mean of the mentioned percentages. CD44 expression was quantified by two independent experienced pathologists blinded to the patient’s diagnosis and outcome, and the mean expression was computed for each slide.

All data were analyzed using IBM SPSS version 23.0 software (IBM Corp., Armonk, NY, USA). Descriptive analysis included mean, standard deviation, median, and range. Data distribution was assessed by the Kolmogorov-Smirnov test, and it demonstrated that all variables did not have a normal distribution (p < 0.05). Thus, nonparametric statistical tests were used for analysis. Differences in age and CD44 expression between the BPH and PAC groups were analyzed through the Mann-Whitney U test. As Gleason scores and grade group are ordinal variables, the total CD44-positive percentage was compared among these groups through the Kruskal-Wallis test. The correlations of total Gleason score, Gleason grade groups, and age with total CD44 expression were analyzed by Spearman’s Rho test. Furthermore, ordinal logistic regression analysis was performed to further assess the relationship between total CD44 expression and total Gleason score and Gleason grade groups. This statistical method was chosen because both the total Gleason score and grade groups are ordinal variables, which allows for the modeling of the probability of each category while accounting for the ordered nature of these outcomes. We considered age as a potential confounder in regression analysis. The results of the logistic regression are expressed in terms of odds ratios (ORs). If the OR is greater than one, it indicates a positive relationship, suggesting that an increase in the predictor variable is associated with higher odds of the outcome occurring. Conversely, if the OR is less than one, it indicates a negative relationship, meaning that an increase in the predictor variable is associated with lower odds of the outcome occurring. The p < 0.05 was considered significant.

This study included 80 PAC and 83 BPH samples. The mean age of the PAC and BPH groups was 70.72 ± 8.86 and 70.54 ± 9.29, respectively. No significant difference was shown between the ages of the two groups (p = 0.967). The mean expression of CD44 in PAC samples was significantly lower than in BPH samples (28.59 ± 14.84 vs. 47.82 ± 14.65, p < 0.001). The details of sample characteristics are shown in Table 1.

The frequency of each score and grade and its mean expression of CD44 based on primary, secondary, and total Gleason scores and grade groups are presented in Table 2. The percentage of CD44 expression in total Gleason scores and Gleason grade groups is shown in Figure 1. The CD44 expression was found to have a significant relationship with the primary, secondary, and total Gleason scores, as well as the grade groups of PAC (p < 0.001). The distribution of CD44 expression according to age of the patients is shown in Figure 2.

There was no significant correlation between the CD44 expression and the age of the PAC samples (r = +0.164, p = 0.146), while a moderate to strong significant negative correlation was found between CD44 expression and primary, secondary, and total Gleason scores and grade groups, as shown in Table 3 (r: −0.543, p < 0.001; r: −0.624, p < 0.001; r: −0.743, p < 0.001; r: −0.732, p < 0.001, respectively). According to ordinal logistic regression analysis, CD44 expression had a significant negative association with total Gleason scores (OR = 0.884, 95% confidence interval = 0.846–0.924, p < 0.001) and grade groups (OR = 0.884, 95% confidence interval = 0.851–0.919, p < 0.001), as shown in Table 4.

The results reported here present a comprehensive analysis of the expression of CD44 biomarker in PAC and BPH samples. Our study included a total of 163 samples, with 80 PAC and 83 BPH samples. The key finding of this study is that the mean expression of CD44 was significantly lower in PAC samples compared to the BPH samples. This suggests that decreased CD44 expression may be associated with the development or progression of prostate cancer. The results further demonstrate a significant negative correlation between CD44 expression and the primary, secondary, and total Gleason scores, as well as the grade groups, in the PAC samples. This indicates that lower CD44 expression is associated with higher Gleason scores and more advanced grade groups, which are indicators of more aggressive and poorly differentiated prostate tumors. The ordinal logistic regression analysis confirmed the significant negative association between CD44 expression and total Gleason scores and grade groups. These findings suggest that CD44 expression may serve as a potential biomarker for prostate cancer aggressiveness and progression. Patients with lower CD44 expression in their prostate tumor samples may have a higher risk of more advanced and aggressive disease, which could have important implications for clinical management and treatment decisions.

Our regression results indicated that with a one-unit increase in CD44 expression, the odds of having a higher total Gleason score or Gleason group decreased by 0.884 times. An OR of less than 1 indicates a negative association, meaning that higher CD44 expression was associated with lower odds of having more advanced disease. The statistical significance of the OR is determined by the p value. In this study, the p values for both conducted regression between CD44 and total Gleason scores, and CD44 and grade groups were <0.001. This indicates that the observed associations are highly unlikely to have occurred by chance and are statistically significant. These findings suggest that lower CD44 expression is associated with increased odds of having more advanced prostate cancer, as indicated by higher Gleason scores and grade groups. Patients with lower CD44 levels may have a higher likelihood of having more advanced prostate cancer, which could inform treatment decisions and guide more intensive monitoring and management strategies. Additionally, integrating CD44 expression into nomograms or risk stratification tools could improve prognosis accuracy and facilitate personalized treatment plans.

A number of previous reports have evaluated the expression of CD44 in the prostate. Some of the findings in those reports were similar to the result presented in our study but others found different results. However, there were only a small number of studies which compared the CD44 expression of PAC with BPH. Thapliyal et al. [17] conducted a study involving 30 PAC samples and 29 BPH samples, revealing that positive CD44 expression was significantly higher in the younger age group. Furthermore, specimens displaying positive CD44 expression were significantly associated with histological diagnoses of BPH compared to those with negative CD44 expression. Notably, cases with CD44 positivity did not exhibit biochemical recurrence, whereas 44.4% of CD44 negative cases showed recurrence. Similarly, in a study by Balcı and Özdemir [18] involving 60 PAC samples and 20 BPH samples, CD44 expression was significantly higher in BPH cases compared to those with PAC. Furthermore, the level of CD44 expression gradually decreased with the increase in histologic grade, although this trend was not statistically significant. The findings from both Thapliyal et al. [17] and Balcı and Özdemir [18] studies were similar to our study but our study included a larger sample size. Other studies examined CD44 expression only in the PAC cases. In the study conducted on 57 PAC samples by Damarasingu et al. [15], no relationship was revealed between age and CD44 expression but a significant reduction in CD44 expression was observed in samples with higher Gleason grades. Additionally, in a study by Imran et al. [13] involving 50 PAC samples, a significant inverse relationship between CD44 expression and Gleason grade score was observed, while age did not exhibit any significant correlation with CD44 expression. In a study done by Kalantari et al. [19], statistical analysis indicated a significant inverse correlation between CD44 expression and Gleason score in PAC. Similarly, in a study by Hirth et al. [20] encompassing 69 PAC samples, a significant association was found between CD44 negativity and higher Gleason scores. Moreover, research by Noordzij et al. [14] involving 97 PAC samples revealed an inverse correlation between CD44 expression with the Gleason score and Gleason grade. Also, it showed that loss of CD44 expression predicted a poor prognosis independent of stage and grade. Additionally, Kallakury et al. [21] found reduced CD44 protein expression in PAC, which correlated with higher tumor grade. Likewise, the research by De Marzo et al. [22] demonstrated a reduction in CD44 expression was correlated with an increase in Gleason grade. As in the studies mentioned above, the CD44 expression alterations in the different grades and scores of PAC were similar to our study. Nevertheless, the Gleason score was not significantly associated with any CD44 isoform in a study by Moura et al. [23]. This may be due to different methodologies used in their study utilizing real-time q-PCR to analyze expression levels of CD44s and its variants.

CD44 expression in other malignancies has also been studied. For example, in colorectal cancer, CD44 expression showed an inverse relationship with lower pathological stages and lymphatic invasion [24]. Another study revealed that higher levels of CD44 expression were significantly associated with the grades of squamous cell lung carcinomas [25]. In addition, lower CD44 expression was more frequently observed in well-differentiated lung adenocarcinomas [25]. Conversely, some findings suggest that elevated CD44 expression may contribute to the aggressiveness and metastatic potential of melanoma cells. A positive correlation was identified between CD44 expression and recurrence in melanoma cases [26]. Furthermore, higher levels of CD44 expression were frequently observed in larger melanoma samples [26]. The seemingly conflicting perspectives may stem from different studies focusing on various tumor types. It is possible that CD44 activation is critical for the progression of tumors originating from certain organs, while tumors from other sources may not rely on such activation [11].

CD44 is a transmembrane glycoprotein that is essential for numerous cellular processes including cell adhesion, proliferation, differentiation, migration, and survival [27]. These functions are vital for normal cells in physiological conditions. The primary function of CD44 is to interact with hyaluronic acid and a variety of stromal ligands, including glycosaminoglycan hyaluronan, collagens, and matrix metalloproteinases. The absence of CD44 facilitates the detachment of cells from the basement membrane, thereby allowing malignant cells to dissociate from their primary location [18]. This process occurs through the degradation of the extracellular matrix, leading to increased cellular motility and invasiveness. Consequently, CD44 can play a crucial role in maintaining the tumor in a differentiated, gland-forming state. This may explain our findings on the inverse correlation of CD44 expression with Gleason scores and grade groups. Additionally, different isoforms or changes in the expression of CD44 have been linked to specific types of cancer [27]. This biomarker functions as a suppressor gene for metastatic tumors in PAC [17]. The progression of PAC is marked by a reduction in both mRNA and protein levels of CD44, leading to elevated tumor grade, metastasis, and aneuploidy [14]. The extracellular domain of CD44 is able to interact with various molecules in the microenvironment, including growth factors and components of the extracellular matrix [28]. Also, the interaction between runt-related transcription factor 2 (RUNX-2) and CD44-ICD (intracellular domain) results in the formation of a complex that triggers the expression of genes linked to metastasis [29]. Due to these interactions, CD44 is believed to play a significant role in regulating cancer progression. It has been revealed that methylation of cytosine within the promoter region of CD44 is correlated with diminished CD44 expression in the human prostate [30]. The upregulation of CD44 expression through specific microRNAs could potentially be employed as an innovative therapeutic approach against PAC. Timely upregulation of CD44 expression may enable the prevention of PAC progression to high grades [18]. The suppression of CD44 in tumors could lead to the degradation of the basement membrane and disruption of stromal-epithelial communication [17]. According to the findings of a study conducted by Iczkowski [12], CD44 exhibits potential as a target for gene therapy through stable reexpression. Based on the abovementioned processes, therapeutic strategies can be envisaged which may include upregulating CD44 expression in PAC to enhance the treatment efficacy. These efforts could lead to innovative treatments and improved outcomes for patients with cancers such as PAC.

This study has several potential limitations which are as follows: (1) it was conducted at a single center, rather than multiple sites. (2) The sample size was relatively small. Larger, multicenter studies are needed to validate these findings. (3) Only a single biomarker (CD44) was evaluated. (4) Only one laboratory method was used to detect CD44 expression. (5) There were no follow-up data included in the study. Future studies could consider conducting prospective longitudinal studies with follow-up data correlating CD44 expression and clinical outcomes, such as prostate-specific antigen levels, tumor stages, progression-free survival, and overall survival to better understand the long-term implications of CD44 expression. Integrating CD44 expression data with other molecular markers may confer more diagnostic and/or prognostic value. Furthermore, exploring CD44-targeted therapies could open new avenues for PAC treatment.

This study provides evidence regarding the role of CD44 expression as a potential biomarker in PAC compared to BPH. Our analysis revealed no significant age differences between the two groups. The most notable finding is the significantly lower mean expression of CD44 in PAC samples compared to BPH samples, suggesting that decreased CD44 expression may be associated with the onset and progression of prostate cancer. Furthermore, the study demonstrated a strong negative correlation between CD44 expression and Gleason scores and grades, indicating that lower levels of CD44 are linked to more aggressive and poorly differentiated tumors. The ordinal logistic regression analysis reinforced these findings, highlighting that higher CD44 expression correlates with lower odds of advanced disease. These results have significant clinical implications, suggesting that CD44 expression could be utilized to stratify patients based on their risk of disease progression. Patients exhibiting lower CD44 levels may require closer monitoring and more aggressive treatment strategies, while those with higher expression may be candidates for less intensive management. Incorporating CD44 expression into nomograms or risk stratification tools may improve the precision of prognosis predictions and support the development of personalized treatment plans, contingent on validation through larger and prospective studies. Overall, our findings advocate for future longitudinal studies with follow-up data to solidify CD44’s prognostic value when integrated with other markers, thus contributing to a more robust prognostic model.

We would like to thank the Vice Chancellor for Research of Shahed University for approving and financially supporting this research.

The study protocol was approved by the Ethics Committee of Shahed University (IR.SHAHED.REC.1402.104).

On behalf of all authors, the corresponding author states that there is no conflict of interest.

This work was funded by Shahed University.

Joben Kianparsa and Masood Soltanipur extracted data and contributed to IHC and H&E staining on the included samples. Also, statistical analysis was done by Masood Soltanipur. The initial draft was written by Masood Soltanipur and Joben Kianparsa, which was finalized by Mohammadreza Jalali Nadoushan. Joben Kianparsa and Masood Soltanipur are equally co-first authors.

Joben Kianparsa and Masood Soltanipur contributed equally as co-first authors.

Data are available if requested from the corresponding author.