The association of the second-to-fourth digit ratio with prostate cancer diagnosed by magnetic resonance imaging-transrectal ultrasound fusion biopsy: A comparative analytical cross-sectional analysis of prospectively recorded data

Abstract

Only a few studies that employed conventional transrectal ultrasound (TRUS) biopsy have investigated the connection between the second digit (2D)-to-fourth digit (4D) ratio and prostate cancer (PCa), and their findings have been conflicting. We aimed to investigate the correlation between the 2D:4D ratio and prostate cancer, identified through a multiparametric magnetic resonance imaging-TRUS fusion biopsy, and its association with clinically significant PCa (csPCa). Patients who underwent multiparametric magnetic resonance imaging/TRUS fusion biopsy due to the prostate imaging reporting and data system 3/4/5 lesions between 2020 and 2024 were included in the study. The patients were divided into 3 groups: study group (n = 168): prostate cancer; study subgroup: csPCa (n = 95); and control group (n = 360): non-cancer. The groups were compared in terms of demographic data, patient characteristics, MRI characteristics, pathological features, cancer stages, and the right-hand 2D:4D ratio. There was a significant difference between the study and control groups for total prostate-specific antigen (PSA) levels, the rate of positive digital rectal examination findings, PSA density, and prostate volume, all favoring the study group (<.001, <.001, <.001, and <.001, respectively). The study and control groups did not show any difference in terms of right-hand 2D:4D ratio. Similarly, no difference was observed between the study subgroup and the control group when the study group was evaluated specifically for csPCa. In a multivariable regression analysis, age and PSA were found to be independent risk factors; prostate volume and the 2D:4D ratio were not significant after adjustment for other variables. The 2D:4D ratio may not be a dependable predictor for both overall and csPCa risk. Considering the literature data and the results of our study, there is an unclear relationship between the 2D:4D ratio and PCa.

1. Introduction

The second (2D)-to-fourth (4D) digit ratio is proposed as a possible biomarker for prenatal exposure to sex hormones like testosterone and estrogen, measuring the length ratio between the second and fourth fingers.^[1] It is widely acknowledged that testosterone and androgen receptors (ARs) play a significant role in the development of prostatic tissue and the initiation of prostate cancer. Many studies have shown a clear link between the number of CAG repeats in the androgen receptor (AR) gene and the length of fingers. Furthermore, there is a concern that individuals with a lower number of AR CAG repeats may have an increased susceptibility to prostate cancer.^[2]

Currently, there have been limited investigations into the correlation between the 2D:4D ratio and prostate cancer, and the results have been inconsistent. Furthermore, all these studies employed conventional transrectal ultrasound (TRUS) biopsy.^[3] The multicenter randomized controlled PRECISION trial demonstrated that the rate of detecting prostate cancer with an International Society of Urological Pathology (ISUP) grade ≥2 was significantly higher in men who underwent multiparametric magnetic resonance imaging (mp-MRI)-TRUS fusion prostate biopsy compared to those who underwent conventional systemic biopsy.^[4] An mp-MRI-TRUS fusion prostate biopsy study will more clearly reveal the existence of a possible relationship.

We hypothesized that the 2D:4D ratio may be an indicator for the diagnosis and severity of prostate cancer. In this context, we aimed to investigate the potential correlation between prostate cancer, identified through a multiparametric magnetic resonance imaging (mp-MRI)-TRUS fusion prostate biopsy, and the 2D:4D ratio, as well as its association with clinically significant prostate cancer (csPCa).

2. Materials and methods

2.1. Compliance with ethical standards

The current study was approved by both the Institutional Medical Ethics Committee of Haseki Training and Research Hospital (May 23, 2024; Approval no. 24-2024) and the Institutional Education Planning Board (Approval no. 137). Prior to performing biopsies, all patients provided verbal and written consent (signed documents are available in the patient files of the hospital).

2.2. Study design

This comparative analytical cross-sectional study, a form of observational study, included a retrospective analysis of data collected prospectively in a tertiary center. The study’s data was obtained from our clinic’s “fusion biopsy data recording system.” However, the data prior to 2019 was not included in the study, as the radiology department switched to Prostate Imaging Reporting and Data System version 2.1 in 2019.^[5] Patients underwent Mp-MRI due to elevated prostate-specific antigen (PSA; >2.5 ng/mL) and/or suspicious digital rectal examination (DRE), and among these, patients who underwent mp-MRI/TRUS fusion prostate biopsy due to the PIRADS 3/4/5 lesions that were accepted as significant lesions between January 2020 and April 2024 were included in the study (n = 616). The patients undergoing active surveillance for prostate cancer (n = 32), as well as those with a prior history of hand surgery or trauma (n = 6), missed data (n = 50) and were excluded from the study. The remaining patients were divided into 2 groups according to the biopsy results: study group (n = 168): prostate cancer; study subgroup: csPCa (n = 95); and control group (n = 360): non-cancer (atypical small acinar proliferation; ASAP, high-grade prostatic intraepithelial neoplasia; HGPIN, prostatitis, benign prostate hyperplasia; BPH; Fig. 1). The groups were compared in terms of demographic data (age, family history, body mass index, comorbidities), patient characteristics (total PSA, PSA density, previous biopsy history, DRE), MRI characteristics (prostate volume, PIRADS score, lesion size, lesion number), pathological features (number of cores, gleason score, ISUP grade), cancer stages, and right-hand 2D:4D ratio. The primary endpoint was to determine the correlation with prostate cancer, while the secondary endpoint was to assess the relationship with csPCa.

Figure 1.

Flowchart of the study.

2.3. Digit ratio

All finger length measurements were conducted by a single experienced observer using a digital vernier caliper to minimize inter-observer variability. The lengths of 2D and 4D were measured from the distal fingertip to the midpoint of each proximal crease (Fig. 2). The 2D:4D ratio was determined by dividing the length of the 2D by the length of the 4D.

Figure 2.

Measurement of 2D and 4D finger lengths by digital vernier caliper, (A) the length of the right hand 2D finger: 66.8 mm, (B) the length of the right hand 4D finger: 65.2 mm, 2D:4D ratio: 1.02. 2D = second digit, 4D = fourth digit.

2.4. Prostate multiparametric magnetic resonance imaging

All mp-MRIs were conducted using either 1.5 Tesla or 3 Tesla MRI machines. The evaluated MRI sequences included T1-weighted, T2-weighted, diffusion-weighted, and dynamic gadolinium-contrast images. An experienced radiologist (XXXX) specializing in prostate MRI examined each mp-MRI image according to the parameters outlined in PIRADS v2.1 guidelines. The radiologist also determined the prostate volumes by utilizing axial and sagittal T2-weighted scans, employing the formula height × breadth × depth/2. The radiologist was blinded to the finger measurements of the patients.

2.5. Biopsy procedure

Patients with PIRADS 3 or more lesions underwent an mp-MRI/TRUS fusion prostate biopsy (target + systemic sampling) via transrectal technique. Fosfomycin (3 g, orally 12 hours before and 24 hours after the biopsy) and a cleansing fleet enema were administered to all patients. Biopsies were performed in the lithotomy position using the Biojet rigid image-fusion platform (D&K Technologies GmbH, Barum, Germany). Pudendal and periprostatic nerve blocks were used for local anesthesia. All suspected lesions received a minimum of 3 core samples, at least 2 from the lesion center (Fig. 3). The systematic biopsy plan involved 10 to 12 peripheral prostate samples.

Figure 3.

Sagittal plane TRUS image that was fused with an mp-MRI image segmented by drawing prostate borders and lesion borders; target lesion sampling (Mp-MRI/TRUS fusion targeted biopsy was performed in this patient who has a 9 mm PIRADS 4 lesion located in the right peripheral zone). mp-MRI = multiparametric magnetic resonance imaging, PIRADS = Prostate Imaging Reporting and Data System, TRUS = transrectal ultrasound.

2.6. Pathologic evaluation

The formalin-fixed, paraffinized tissues were examined for pathological analysis. The tissues were appraised as 4 µm thick sections and stained with Hematoxylin&Eosin (Fig. 4A). The cases were evaluated by evaluating the primary and secondary Gleason patterns and categorized according to the ISUP 2014 classification (Fig. 4B, C).^[6] Immunohistochemical analysis was employed to discriminate well-differentiated tumors from nonneoplastic glands by observing the absence of basal layer markers such as HMWCK, p63, and SMA. Immunohistochemical markers, including Nkx3, PSA, and PSAP, were employed in the assessment of poorly differentiated tumors.

Figure 4.

Small magnification, an appearance of infiltrated prostate stroma, and tumoral infiltration with an increased gland/stroma ratio (H&E, ×100) (A), higher magnification, non-tumoral prostatic acini with darker and narrower cytoplasm, Gleason pattern 3 tumor cells with clearer cytoplasm, which are widely distributed among the glands, have lost their myoepithelial layer, and form simple tubular structures according to the pattern structure, and Gleason pattern 4 tumor cells that form complex structures that are less abundant than pattern 3 (H&E, ×200) (B), tumoral infiltration with solid growth, hyperchromatic nuclei, and narrow cytoplasm (C).

2.7. Sample size

The study will be conducted retrospectively, incorporating all eligible patients from our clinic’s data pool to accurately represent the general population. In this context, the study, with a meticulously defined study and control group, included a total of 528 patients selected according to specified inclusion and exclusion criteria from 2020 to 2024.

2.8. Statistical analysis

Statistical analysis of the study was performed using IBM SPSS Statistics® version 29.0.2 software (IBM Corp., Armonk). Numerical variables that conform to a normal distribution were displayed using the mean ± standard deviation (SD), whereas numerical variables with a distribution that deviates from normality were displayed using the median and minimum (min.)–maximum (max.) values. Categorical variables were presented as a proportion represented by a percentage symbol. The normality of the quantitative variables was assessed using histogram graphics, coefficient of variation, skewness and kurtosis values, normal Q–Q plot and detrended normal Q–Q plot graphics, and the Kolmogorov–Smirnov test.The Independent-Samples T test was used to compare numerical variables (age and body mass index) that exhibited a normal distribution between 2 independent groups. On the other hand, the Mann–Whitney U test was used to compare variables that did not exhibit a normal distribution (total PSA, PSA density, prostate volume, number of significant lesions on the mp-MRI, total number of cores in biopsy, and right-hand 2D:4D ratios). The Chi-square test was used to compare categorical variables (comorbidity, DRE findings) between groups. Cohen’s d (for t tests), Mann–Whitney U effect size (r); (for non-parametrics), and Cramer’s V (for the Chi-square tests) have now been calculated and reported alongside P-values in Tables 1–4. P-values of .05 or below were considered statistically significant at a 95% confidence level.

Table 1.

Comparison of demographic characteristics and comorbidities.

Groups Variables	Study group (overall prostate cancer) (n = 160)	Control group (non-cancer) (n = 368)	P value	Effect size (95% CI)
Age (yr), mean ± SD	63.45 ± 6.16	60.80 ± 6.48	<.001 *	d = 0.42
BMI (kg/m²), mean ± SD	27.62 ± 3.47	27.46 ± 3.49	.61*	d = 0.047
Comorbidity, n (%)			.076†	OR = 1.63
Metabolic syndrome (+)	31 (18.5%)	44 (12.2%)
Metabolic syndrome (−)	129 (81.5%)	324 (87.8%)

Statistically significant value (P < .05) is indicated in bold.

BMI = body mass index, SD = standard deviation.

^*Independent-samples t test.

^†Chi-square test (continuity correction).

Table 4.

Comparison of right hand 2D:4D ratio between the csPCa patient group and control group.

Groups Variable	Study subgroup (csPCa) (n = 95)	Control group (non-cancer) (n = 368)		P value	Effect size (95% CI)
Right hand 2D:4D ratio, median (min.–max.)	0.96 (0.86–1.09)		0.96 (0.71–1.14)	.78*	r = 0.017

2D:4D = second digit-to-fourth digit, csPCa = clinically significant prostate cancer.

^*Independent-samples Mann–Whitney U test.

A multivariable regression analysis was conducted to account for any confounding variables that may affect the 2D:4D ratio. Age, total PSA level, and prostate volume were included as factors in the model for evaluating the relationship between the right-hand 2D:4D ratio and prostate cancer outcomes. This adjustment aimed to minimize bias from these variables, which are known to be associated with both PCa risk and potentially with digit ratio.

3. Results

The mean age was significantly higher in the study group compared to the control group (Cohen’s d = 0.41, 95% CI: 0.26–0.56). Groups were similar in terms of major comorbidities and metabolic syndrome (OR = 1.63, 95% CI: 0.98–2.68; Table 1).

There was a significant difference between the study and control groups for total PSA levels (R = 0.15, 95% CI: 0.146–0.149), the rate of positive DRE findings (OR = 5.86, 95% CI: 3.42–10.05), PSA density (R = 0.285, 95% CI: 0.283–0.287), and prostate volume (R = 0.218, 95% CI: 0.216–0.219), all favoring the study group. There was no significant difference in terms of the number of PIRADS 3/4/5 lesions detected on the mp-MRI (R = 0.002, 95% CI: 0.0003–0.0036) and the total number of cores taken in biopsy (R = 0.093, 95% CI: 0.092–0.095; Table 2).

Table 2.

Comparison of PSA derivates, prostatic lesion characteristics, and biopsy features between the groups.

Groups Variables	Study group (overall prostate cancer) (n = 160)	Control group (non-cancer) (n = 368)	P value	Effect size (95% CI)
Total PSA, median (min.–max.)	7.4 (2.4–290)	6.3 (1.4–39.2)	<.001 *	r = 0.15
DRE findings, n (%)			<.001 †	OR = 5.86
Positive	48 (28.6%)	23 (6.4%)
Negative	120 (71.4%)	337 (93.6%)
PSA density, median (min.–max.)	0.16 (0.04–5.63)	0.11 (0.02–1.21)	<.001 *	r = 0.285
Prostate volume, median (min.–max.)	46 (9–275)	57.5 (12–209)	<.001 *	r = 0.218
Number of significant lesions on the mp-MRI, median (min.–max.)	1 (1–3)	1 (1–4)	.95*	r = 0.002
Total number of cores in biopsy, median (min.–max.)	17 (8–24)	17 (5–22)	.029*	r = 0.093
csPCa, n (%)	96 (60%)	0	NA

Statistically significant values (P < .05) are indicated in bold.

csPCa = clinically significant prostate cancer, DRE = digital rectal examination, mp-MRI = multiparametric magnetic resonance imaging, NA = not available, PSA = prostate specific antigen.

^*Independent-samples Mann–Whitney U test.

^†Chi-square test (continuity correction).

The study and control groups did not show any significant difference in terms of right-hand 2D:4D ratios (R = 0.055, 95% CI: 0.053–0.056; Table 3; Fig. 5).

Table 3.

Comparison of right hand 2D:4D ratio between the groups.

Groups Variable	Study group (Overall prostate cancer) (n = 160)	Control group (non-cancer) (n = 368)	P value		Effect size (95% CI)
Right hand 2D:4D ratio, median (min.–max.)	0.96 (0.86–1.11)	0.96 (0.71–1.14)		.222*	r = 0.055

2D:4D = second digit-to-fourth digit.

^*Independent-samples Mann–Whitney U test.

Figure 5.

Right hand 2D:4D ratios between study and control groups. 2D = second digit, 4D = fourth digit.

Similarly, when the study group was evaluated specifically for csPCa instead of overall prostate cancer, no significant difference was observed compared to the control group (R = 0.017, 95% CI: 0.015–0.019; Table 4; Fig. 6).

Figure 6.

Right hand 2D:4D ratios between study subgroup and control groups. 2D = second digit, 4D = fourth digit.

In a multivariable regression analysis, age, PSA, prostate volume, and right-hand 2D:4D ratio were included in the model, and the association of each variable with prostate cancer was analyzed independently of the others. Age and PSA were found to be independent risk factors; prostate volume and the 2D:4D ratio were not significant after adjustment for other variables (Table 5).

Table 5.

Multivariable regression analysis.

Parameters Variables	Beta coefficient (b)	P value	Adjusted OR (95% CI)
Age	0.032	.003	1.03 (1.01–1.05)
PSA level	0.185	.001	1.20 (1.08–1.33)
Prostate volume	−0.014	.112	0.99 (0.97–1.01)
Right-hand 2D:4D ratio	0.221	.257	1.25 (0.87–1.78)

2D:4D = second digit-to-fourth digit.

4. Discussion

Our study’s findings indicate that there is no substantial association between the 2D:4D ratio and prostate cancer risk (both overall prostate cancer and csPCa). This finding contributes to the ongoing debate about the possible association between the 2D:4D ratio and prostate cancer, providing additional evidence to question the reliability of the 2D:4D ratio as a biomarker for the risk of prostate cancer. Consistent with prior research indicating that the right-hand 2D:4D ratio exhibits more pronounced sex differences and serves as a more reliable measure of prenatal androgen exposure than the left hand only right-hand measurements were analyzed to maintain methodological consistency and facilitate comparison with earlier studies.^[2]

A meta-analysis study examining the correlation between the 2D:4D ratio and various types of cancer found that a low 2D:4D ratio was associated with prostate cancer, gastric cancer, and brain tumors. Conversely, a high 2D:4D ratio was associated with an increased risk of breast cancer and cervical dysplasia. The study also reported that the 2D:4D ratio was not linked to the stage of prostate, breast, or gastric cancer.^[7] This comprehensive meta-analysis examined a total of 9 studies (8 case-control or cross-sectional studies, involving 4128 patients, and 1 prospective cohort study, involving 6458 men from a community sample) on prostate cancer up until 2018. The majority of the selected studies indicated that prostate cancer was linked to a lower 2D:4D ratio. However, 2 groups either found no association between the 2D:4D ratio and prostate cancer or reported that a higher 2D:4D ratio was associated with prostate cancer. The majority of studies included in the meta-analysis did not establish a correlation between the 2D:4D ratio and important clinical parameters such as Gleason score, presence of metastasis, family history of prostate cancer, and age at diagnosis. A single study observed a negative correlation between the 2D:4D ratio and both the amount of core cancer volume and the number of biopsy cores with a high Gleason grade. The disparate findings from all these different studies highlight the intricacy of establishing a definitive correlation between the 2D:4D ratio and prostate cancer. This is likely the result of variations in study designs, sample sizes, demographic characteristics, age, and measuring methodologies.

Our study was conducted on a Turkish community, and it is worth noting that the absence of an association between the 2D:4D ratio and prostate cancer in our dataset may be specific to this particular demographic. The studies that demonstrated an association between the 2D:4D ratio and prostate cancer were undertaken in several geographical locations, including the United States of America,^[7,8] Spain,^[9] Brazil,^[10] Korea,^[11] and the United Kingdom.^[12] On the other hand, the research conducted in Australia^[13] and a separate study conducted in the Brazilian population,^[14] which likewise did not uncover any significant connection, were more in line with our own findings. In addition to these racial or ethnic differences and ambiguity, there is also uncertainty regarding the specific rates and cutoff values linked to prostate cancer. The research carried out in Spain revealed that people with a 2D:4D ratio over 0.95 were linked to the development of prostatic neoplasia. Conversely, the study conducted in Korea indicated that those with a 2D:4D ratio below 0.95 were more likely to have prostate cancer in biopsy. All these findings emphasize the necessity for more investigation to examine the potential influence of racial and geographical disparities on the correlation between the 2D:4D ratio and the risk of prostate cancer. Such investigations could provide insight into whether the 2D:4D ratio may be a more significant indicator in certain populations compared to others.

Population demographics vary significantly. Research arises from various genetic backgrounds (e.g., Caucasian, Korean, Brazilian) and environments characterized by differing lifestyle factors (diet, environmental exposures, healthcare access) that may independently affect PCa risk and potentially alter any biological association with prenatal androgen exposure as indicated by the 2D:4D ratio. The Turkish population examined in this study may possess unique genetic polymorphisms associated with androgen metabolism or receptor sensitivity, differing from those in populations identified in positive studies.

Another important reason for the difference between studies may be measurement techniques. We used direct measurement conducted by a single experienced observer with digital vernier calipers, whereas other studies have employed radiographs,^[7,8,12] photocopies,^[9] or self-measurement.^[13] The methods exhibit variations in precision and susceptibility to bias; radiographic measurements may provide more accurate assessments of bone length but entail radiation exposure, whereas direct measurements or photocopies could be affected by skin creases or the technique of the observer.

This study also controlled for potential confounding variables, including age, PSA level, and prostate volume, using multivariable regression analysis. Modifications for these variables did not significantly affect the outcomes, and the correlation between the right-hand 2D:4D ratio and PCa remained statistically nonsignificant. This conclusion indicates that the observed lack of correlation is probably not primarily attributable to these confounding factors, but other unmeasured variables, such as hormone status and genetic factors, may still be influential.

It is crucial to assess our study, taking into account both its limitations and strengths, in order to properly interpret the results. Our study’s primary limitations are the use of data from a single center and its retrospective design. However, the prospective data collection mitigates this limitation. Another limitation was that we did not analyze finger lengths according to age groups. In this context, an alternative study design has the potential to reveal meaningful differences among various age groups. Another limitation is that only the right-hand 2D:4D ratio was measured. Although this choice was based on previous evidence indicating stronger associations with prenatal androgen exposure, future studies assessing both hands could provide a more comprehensive evaluation. Another important limitation is that the study was conducted only on the Turkish population and did not include hormonal and genetic factors, which may impact the interpretation of the results. Despite these limitations, the significant patient population and the utilization of MR-Trus fusion biopsy, a contemporary technology, are the primary strengths that set our study apart from previous research in the field. Therefore, our research can make a valuable contribution to the existing body of literature.

5. Conclusions

Our study contributes to the existing evidence indicating that the right-hand 2D:4D ratio may not be a dependable predictor for both overall prostate cancer and csPCa risk. Considering the literature data and the results of our study, there is an unclear relationship between the right-hand 2D:4D ratio and prostate cancer. This emphasizes the necessity for further research to explore different new biomarkers and risk factors.

Author contributions

Conceptualization: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli, Arda Meric, Rustu Turkay.

Data curation: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli, Arda Meric, Rustu Turkay.

Formal analysis: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli, Arda Meric, Rustu Turkay.

Investigation: Akif Erbin.

Methodology: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli, Arda Meric, Rustu Turkay.

Supervision: Akif Erbin, Rustu Turkay.

Visualization: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli.

Writing – original draft: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli, Arda Meric.

Writing – review & editing: Akif Erbin, Caglar Dizdaroglu, Feyzi Sinan Erdal, Sami Sekkeli, Rustu Turkay.