Intrauterine fetal growth restriction (FGR) affects approximately 3-8% of pregnancies and is a leading cause of perinatal morbidity and mortality. It is primarily associated with placental insufficiency and impaired placental perfusion, which can lead to chronic fetal hypoxia [1].

FGR is defined as the failure of the fetus to reach its genetically predetermined growth potential and is typically identified when the estimated fetal weight falls below the 10th percentile for gestational age and sex [1,2]. Approximately 70% of fetuses below the 10th percentile are constitutionally small but healthy, while approximately 30% are truly growth-restricted due to pathological causes [3]. Of these, 20% result from placental insufficiency, and 10% are linked to chromosomal abnormalities, infections, or teratogenic exposure [4,5]. Furthermore, advanced maternal age increases the likelihood of conditions, such as obesity, diabetes, and hypertension, contributing to adverse outcomes, such as preterm birth and low birth weight [6]. Early-onset FGR (before 32 weeks) is often associated with chromosomal abnormalities or severe placental insufficiency and carries a higher risk of adverse outcomes, including perinatal death. Late-onset FGR, which is more frequent and often less severe, may present with subtler placental histological abnormalities [7-9]. Pathological placental changes in FGR include maternal vascular malperfusion, infarcts, retroplacental hemorrhage, and accelerated villous maturation, although up to 25% of placentas appear normal on routine histological examination [10,11]. However, histological changes associated with FGR are not specific, as similar lesions can also be observed in pregnancies without growth restriction [12,13]. Moreover, placental weight and volume show stronger correlations with pregnancy outcomes than do individual histological findings [14,15]. The heterogeneity of FGR, the uneven distribution of placental pathology, and the limitations of routine histological sampling further reduce the diagnostic value of morphological assessment [16-18].

The diagnosis of FGR relies on serial ultrasound biometry and Doppler studies, with key markers including estimated fetal weight (based on head circumference, biparietal diameter, abdominal circumference, and femur length) and Doppler indices of the umbilical artery and fetal cerebral circulation [16,17]. Furthermore, low levels of first-trimester biomarkers such as pregnancy-associated plasma protein-A (PAPP-A) ≤0.45 multiples of the median (MoM) [18] and high levels of beta-human chorionic gonadotropin (β-hCG) >3.0 MoM in the second trimester are associated with an increased risk of FGR [19]. However, their sensitivities and specificities are limited for late-onset FGR, which often lacks pronounced biochemical abnormalities [18,19]. Despite these advances, approximately half of FGR cases remain undetected until birth. Current clinical practice lacks a highly sensitive and specific screening test for FGR, highlighting the need for further research on reliable biomarkers and predictive tools [20].

Cripto-1 is a protein belonging to the epidermal growth factor-Crypto/fibroblast growth factor receptor-like 1/cryptic family that plays critical roles in embryogenesis, epithelialmesenchymal transition (EMT), cell migration, and tumor progression [21-23]. During early embryonic development, Cripto-1 is essential for germ layer differentiation and organogenesis and is minimally expressed in adult tissues [23]. In placental biology, trophoblasts share certain characteristics with tumor cells, including invasive behavior, which is regulated in part by EMT processes [24]. It has been suggested that aberrant regulation of EMT may underlie placental insufficiency and contribute to the development of FGR [25,26]. Similarly, Bandeira et al. [27] reported that Cripto-1 expression was significantly higher in cases of placenta accreta than in controls and that expression intensity correlated with invasion depth. Although the role of Cripto-1 in cancer has been widely studied, its role in trophoblast differentiation and placental pathophysiology [28,29], particularly in FGR, remains unclear. Understanding the expression of Cripto-1 in normal and pathological placental tissues could provide insight into the molecular mechanisms underlying abnormal placental development and identify potential biomarkers or therapeutic targets.

The objective of this study was to semi-quantitatively analyze the expression of Cripto-1 protein in placental tissue from term pregnancies complicated by FGR and to compare it with its expression in placentas from healthy term pregnancies. We hypothesized that Cripto-1 expression is significantly reduced in placentas associated with FGR, reflecting impaired trophoblast differentiation and defective EMT regulation. Additionally, we aimed to investigate whether Cripto-1 expression correlates with clinical and morphological indicators of placental insufficiency, including pathological Doppler of the umbilical artery and fetal circulation, altered ultrasound fetal biometry, fetal and placental birth weights, calculated fetoplacental ratios, and maternal serum levels of β-hCG and PAPP-A. We also examined the effect of preeclampsia on these parameters in the FGR group. Furthermore, we examined the impact of maternal age and parity on pregnancy outcomes by comparing FGR cases with uncomplicated pregnancies.

We conducted a retrospective cohort study based on the immunohistochemical analysis of archived human placental tissue samples collected between 2016 and 2020. As the study was retrospective and involved de-identified biological material without direct patient involvement, informed consent was waived by national regulations and institutional policy, with approval from the Institute’s Ethics Committee.

This retrospective cohort study included 153 placentas from term pregnancies (≥37 weeks), 122 of which were complicated by FGR, and 31 gestational age-matched placentas from uncomplicated term singleton pregnancies with normal neonatal outcomes, which served as controls. FGR diagnosis was based on at least two serial ultrasound measurements showing an estimated fetal weight below the 5th percentile for gestational age and sex, confirmed at birth [30]. The exclusion criteria were the same for the FGR and control groups and included multiple gestations, infections, maternal chronic or autoimmune disease, fetal anomalies, and late fetal demise.

Relevant clinical data were retrieved from the electronic birth registry of the Clinical Hospital Center Rijeka, Croatia. These included maternal age, parity, fetal sex, and birth weight (n=153). In the FGR group (n=122) we also collected data, if available, on the presence of preeclampsia, Doppler data of the umbilical artery and fetal circulation, serial ultrasound biometry measurements (biparietal diameter, abdominal circumference, and femur length), fetoplacental ratio, serum levels of PAPP-A obtained in the first trimester, and β-HCG obtained in the second trimester. All women who underwent medical termination of pregnancy (MTOP) during the study period were included in the study, irrespective of whether they attended the scheduled 14-16 days follow-up visit. The MTOP protocols, ultrasound criteria, and clinical management did not change substantially throughout the study period, with no substantive changes in drug regimens or diagnostic standards.

Placental samples were subjected to weight measurements and standardized macroscopic and microscopic pathological examinations [31]. To reduce observer bias, all samples were assessed in a blinded manner, and the recorded macroscopic changes included placental infarction, fibrin deposits, hematoma, and chorionic villi abnormalities.

Tissue microarrays were constructed from formalin-fixed paraffin-embedded blocks. Immunohistochemical staining was performed using the labeled streptavidin-biotin method and Dako TechMate Horizon automated system (Dako North America, Inc., Carpinteria, CA, USA). Antigen retrieval was performed in citrate buffer (pH 6.1), whereas Cripto-1 was pretreated with proteinase K. The anti-Cripto-1 antibody (Abcam ab19917) was used at a 1:100 dilution. Quantification was performed by analyzing the scanned immunohistochemically stained slides, which were assessed semi-quantitatively by two independent observers. Cripto-1 expression was evaluated as the percentage of positively stained cells (PPS) out of 100 counted cells, whereas the staining intensity reaction (IR) was assessed qualitatively, as shown in Fig. 1 and detailed in Table 1. The immunohistochemical index (IHI) was calculated by multiplying the PPS and IR scores, as previously reported [21,25,26]. Furthermore, when the number of observations was small in certain groups, they were combined (e.g., two or more parities). Additionally, IR variables were grouped into two categories as 0, 1=1 and 2, 3=3, similar to the IHI variables, which were grouped into 0, 4=4 and 8, 12=8. This was performed to test for differences in incidence between the control and study groups and later in the sub-analyses within the study group. Therefore, we aimed to ensure sufficient statistical power and avoid unreliable comparisons based on very small cell counts. While regrouping may reduce the sensitivity to detect subtle differences, it was considered necessary to provide a more robust statistical analysis.

Statistical analyses were performed using SPSS software version 23.0 (IBM Corp., Armonk, NY, USA). The distribution of quantitative variables was tested for normality using the Kolmogorov-Smirnov and Shapiro-Wilk tests, whereas the homogeneity of variances was assessed using Levene’s test. Differences between groups for continuous variables were analyzed using the parametric Student’s t-test or nonparametric Mann-Whitney U-test, depending on the data distribution. Categorical variables were compared using the chi-squared test or Fisher’s exact test, as appropriate. Given the exploratory nature of this study and the relatively small sample size, no formal correction for multiple comparisons (e.g., Bonferroni or false discovery rate) was applied. Consequently, P-values were reported without adjustment and should be interpreted with caution, as type I errors cannot be excluded. Statistical significance was set at P<0.05.

Placental tissue and clinical data from term deliveries (2016-2020) were screened. After applying the exclusion criteria, 153 pregnancies were included and categorized as having FGR (n=122) or uncomplicated (n=31), as shown in Fig. 2.

The mean age of patients in the FGR group was significantly higher than that in the control group (P=0.042), whereas no significant difference was observed in fetal gestational age at delivery (P=0.123). However, a significant difference was observed in parity between women with FGR and the control group (P=0.0046). The maternal demographic and clinical characteristics are summarized in Table 2.

Ultrasound and biomarker data were incomplete. In the FGR group, data on ultrasound biometry measurements were successfully collected in 80 patients, with a mean ultrasound-measured biparietal diameter of 89.65±5.76 mm (95% confidence interval [CI], 87.89-90.99), femur length of 69.20±5.87 mm (95% CI, 67.20-70.57), and abdominal circumference of 304.01±23.59 mm (95% CI, 296.7-309.2). Additionally, data from 121 Fp-ratios were collected with a mean value of 351.83±79.81 (95% CI, 330-357.3). However, data on β-hCG and PAPP-A levels were successfully collected in only 34 patients. The mean level of β-hCG was 38.69±21.32 IU/L (95% CI, 28.91-40.64) or 1.14±0.60 MoM (95% CI, 0.863-1.196), while the mean PAPP-A level was 2,664.05±1,496.10 mIU/L (95% CI, 1,897-2,802) or 1.08±0.57 MoM (95% CI, 0.744-1.511).

Cripto-1 protein expression was compared between term placentas from pregnancies with FGR (n=122) and normal pregnancies (n=31).

Cripto-1 expression was analyzed based on the intensity of cellular staining (IR) in both groups. In the FGR group, the most frequently observed intensity was 2 (moderate), which was present in 77 placentas (63.1%). Similarly, in the control group, intensity 2 was the most common, occurring in 24 placentas (77.4%), as shown in Fig. 3. Due to the low number of cases in IR categories 0 and 3, the samples were grouped into two categories. In the FGR group, intensity 2 was present in 111 placentas (91.0%), whereas in the control group, 27 placentas (87.1%) showed the same intensity, with no significant difference between the groups (P=0.516). Cripto-1 expression was assessed using the IHI and categorized into four groups based on its value. In both the FGR and control groups, the most common IHI was 8 (moderate), which was found in 77 FGR placentas (63.1%) and 24 control placentas (77.4%) (Fig. 4). Due to the small number of cases in IHI categories 0 and 12, the samples were grouped into two categories. The majority of FGR placentas IHI 8 (modarete) and control placentas (27 [87.1%]) had an IHI of 8, which was not significantly different (P=0.104). The results are summarized in Table 3. Furthermore, differences in gestational age, biparietal diameter, femur length, abdominal circumference, fetal weight, Fp-ratio, β-hCG, and PAPP-A were explored between participants with lower (IHI=4) and higher (IHI=8) immunohistochemical staining indices. However, no significant differences were observed in any of the measured parameters based on IHI staining. Differences in β-hCG and PAPP-A protein levels were not tested for Cripto-1 expression due to the small number of participants in the group with lower marker values. The results are summarized in Table 4.

Furthermore, in patients with FGR, no statistically significant difference was observed in IR (P=0.423; odds ratio [OR], 0.400; 95% CI, 0.07524-2.239) and IHI (P=0.6657; OR, 1.333; 95% CI, 0.2399-7.982) Cripto-1 protein levels between patients with pathological Doppler findings of the umbilical artery and fetal circulation. Additionally, in the FGR group, no significant correlation was observed between Cripto-1 protein expression (IR and IHI) and the recorded macroscopic changes, including placental infarction, fibrin deposits, hematoma, and abnormalities in the chorionic villi. OR estimates for associations between IR and IHI Cripto-1 expression and placental histopathological changes were unstable because of the absence of low-expression cases. No significant associations were observed (all 95% CI included 1; Table 5).

In the examined group of placentas with FGR, preeclampsia was diagnosed in 27 patients. However, no significant differences were observed between patients with and without preeclampsia regarding maternal parity (P=0.623), fetal gestational age (P=0.706), and fetal sex (P=0.748), or Cripto-1 expression, including IR (P=0.741) and IHI (P=0.675). Likewise, fetal biometric parameters-including biparietal diameter (P=0.280), femur length (P=0.661), and abdominal circumference (P=0.534)-did not differ significantly between the two groups, nor did fetal weight (P=0.933), the Fp ratio (P=0.315), β-hCG levels (P=0.947), or PAPP-A levels (P=0.813).

This study examined the expression of Cripto-1 in term placentas from pregnancies complicated by FGR compared with healthy controls. Contrary to our initial hypothesis, Cripto-1 expression did not differ significantly between the groups, nor was it associated with clinical parameters or histopathological findings. The absence of a difference may reflect the temporal dynamics of Cripto-1, whose expression peaks in early placental development and declines later in gestation, suggesting that its regulatory role is more relevant in earlyonset placental dysfunction than at term [32]. Cripto-1’s established role in EMT and early trophoblast invasion provides a strong biological rationale for its high activity in the first trimester, when successful implantation and placentation require dynamic tissue remodeling [33]. Once placental anchoring and vascularization are stabilized, excessive EMT can be detrimental and potentially lead to uncontrolled invasion [33]. The absence of differential Cripto-1 expression at term in our study may therefore reflect physiological downregulation of this pathway in late pregnancy, when the placental architecture has already matured and invasion is complete. This temporal regulation is consistent with developmental models in which EMT-related signals, including Cripto-1, act transiently to initiate invasion and differentiation but are subsequently silenced to maintain placental homeostasis [33]. From a pathophysiological perspective, this implies that disturbances in Cripto-1 expression may be most relevant in early-onset FGR, in which defective invasion is a key driver [34], whereas its role in late-onset FGR is likely to be minimal [33]. Consistently, a study of placenta increta and placenta previa demonstrated markedly elevated placental Cripto-1 expression compared with that in controls, supporting its role in promoting trophoblast invasion in abnormally adherent placentation, whereas our findings suggest physiological downregulation once placentation is stabilized [35]. Moreover, maternal serum studies have shown a negative correlation between Cripto-1 levels and gestational age (r=−0.325; P<0.001), further supporting the concept that Cripto-1 expression is highest in early gestation but progressively declines as pregnancy advances [36]. Together, these observations provide biological plausibility for our null findings at term and suggest that Cripto-1 may be primarily relevant in early-onset placental dysfunction.

The observed maternal characteristics, such as advanced age and primiparity, in the FGR group are consistent with known risk factors for pregnancy complications, including preeclampsia [37,38]. However, in this cohort, these factors, as well as the presence of preeclampsia in the FGR group, did not translate into altered Cripto-1 expression. This finding is consistent with previous evidence showing that advanced maternal age does not significantly modify the association between birth year and the risk of preeclampsia, with comparable adjusted odds ratios among women aged <35 years and those aged ≥35 years [39]. Based on current knowledge, maternal serum β-hCG and PAPP-A levels in this FGR cohort were within normal limits [18,19] and did not show a strong biochemical signature associated with severe placental dysfunction. This reflects the fact that many of the FGR cases in this cohort may have been late-onset, in which biochemical markers are often less markedly altered than in early-onset disease [18,19]. This also suggests that, in this cohort, serum biomarkers alone may have limited predictive value for FGR severity or the presence of preeclampsia.

This retrospective, single-center design is a major limitation, as it restricts external validity and may limit the generalizability of our findings to broader populations. Additional limitations include missing control data, a modest sample size, and the subjective nature of the immunohistochemical assessment, all of which may have reduced statistical power. Multiple statistical tests were performed without adjustment; therefore, type I errors could not be excluded. Importantly, the analysis was restricted to term placentas, and most FGR cases were late onset, in which the pathological and molecular alterations are often subtle. Therefore, these findings may not apply to early-onset or severe FGR, which are more strongly associated with abnormal trophoblast invasion. Furthermore, as follow-up data were incomplete for some patients, certain outcomes may have been underreported. Future prospective multicenter studies should include both early- and late-onset phenotypes, apply quantitative approaches, and evaluate Cripto-1, along with other EMT-related markers, to clarify its temporal role in placental dysfunction.

In conclusion, this study found no significant differences in Cripto-1 expression between FGR and control placentas or associations with clinical or histopathological parameters. These findings suggest that Cripto-1 plays a critical role in early placental development, with limited relevance at term. Future studies should investigate the temporal dynamics of early-onset and severe FGR and assess Cripto-1 alongside other EMT-related markers to improve early detection of placental dysfunction.