Histogram analysis based on contrast-enhanced T1-weighted imaging in pituitary macroadenomas: relation to histological subtype and expression

Introduction

Pituitary adenomas, originating from the adenohypophysis, account for 16.2% of all primary intracranial tumors (1). Pituitary adenomas are classified into microadenomas (<10 mm) and macroadenomas (>10 mm) (2). Macroadenomas represent about 50% of pituitary adenomas in the clinic, and most patients currently require transsphenoidal surgery intervention. The 2022 World Health Organization (WHO) classification of pituitary tumors employs immunohistochemistry to categorize pituitary adenomas into subtypes. Based on the transcription factors (PIT-1, TPIT, SF-1, GATA-3, and ERα), the adenohypophysis is composed of at least six normal cell types: somatotrophs, lactotrophs, mammosomatotrophs, and thyrotrophs are of PIT-1 lineage, corticotrophs are of TPIT lineage, and gonadotrophs are of SF-1 lineage (3). Gonadotropic cells primarily secrete gonadotropins, including luteinizing hormone (LH) and follicle-stimulating hormone (FSH) (3). Preoperative hormone measurements remain insufficient for the diagnosis of gonadotroph macroadenomas, and the corresponding subtype differentiation still depends on postoperative immunohistopathology. Meanwhile, the previous study indicates that non-gonadotroph macroadenomas are more prone to invade the cavernous sinus than gonadotroph macroadenomas (4). Recent therapeutic advancements have identified somatostatin receptor type 3 in 94% of gonadotroph pituitary adenomas, offering a pharmacological target as an alternative to surgical intervention (5,6). Thus, preoperative differentiation between gonadotroph and non-gonadotroph macroadenomas facilitates optimal treatment planning.

Positron emission tomography-computed tomography (PET-CT) has been validated as an effective diagnostic modality for pituitary adenoma detection (7). Sumida et al. (8) reported that PET-CT demonstrates superior detection efficacy for normal pituitary tissue in macroadenomas, particularly serving as a critical complementary imaging technique when magnetic resonance imaging (MRI) findings are limited. Nevertheless, MRI remains the preferred imaging modality for detecting pituitary and parasellar lesions due to its exceptional spatial resolution and superior diagnostic accuracy. Conventional MRI has been utilized to distinguish between growth hormone-producing and non-growth hormone-producing pituitary adenomas (9). Davies et al. (10) quantitatively estimated tumor volume in 99 pituitary adenoma patients based on conventional MRI, offering imaging guidance for surgical approach selection without subtype characterization. However, conventional MRI falls short in directly quantifying the characteristics of pituitary adenomas, potentially limiting its efficacy in subtype differentiation. Contrast-enhanced T1-weighted imaging (CE-T1WI) histogram analysis translates voxels into histograms to further elucidate tumor homogeneity/heterogeneity (11), thereby providing a more detailed understanding of tissue microstructure and the differentiation of pituitary adenoma subtypes.

The Knosp grading system is widely used to assess pituitary adenoma aggressiveness by classifying parasagittal extension on preoperative coronal MRI (12). Knosp grades 3 and 4 indicate the presence of cavernous sinus infiltration due to invasive growth, whereas Knosp grades 1 or 2 suggest no cavernous sinus infiltration (10). CE-T1WI histogram analysis can effectively quantify the anatomy, vascular supply, and cellular composition of lesions. Combined with Knosp grading, this study aims to precisely characterize adenoma microenvironmental heterogeneity and biological behavior.

The invasiveness of pituitary adenomas is also closely associated with various pathological indicators, among which the expression levels of P53 protein and Ki-67 proliferation index are pivotal markers for assessing tumor biological behavior (13). Ki-67, a marker of cellular proliferation, typically reflects heightened proliferative activity when its index is elevated (14). Pituitary adenomas with high Ki-67 expression levels are often associated with larger tumor volumes and exacerbated clinical symptoms, particularly in cases exhibiting invasive growth or postoperative recurrence. P53, a tumor suppressor, frequently undergoes mutations or inactivation, leading to dysregulated cell cycles and promoting tumor invasiveness and malignant transformation. A study has shown that high expression of P53 in pituitary adenomas is strongly correlated with increased tumor aggressiveness and poorer prognosis (15). The combined assessment of P53 and Ki-67 expression levels offers a more precise prognostic evaluation, aiding clinicians in predicting recurrence risks and malignant potential, thereby facilitating individualized treatment strategies.

Recent study has utilized CE-T1WI histogram analysis to preoperatively predict tumor-infiltrating CD8⁺ T cell levels in glioblastoma patients, revealing that CD8⁺ T cell levels are inversely correlated with histogram metrics such as mean, 5th, 10th, 25th, and 50th percentiles (16). Ouyang et al. (17) demonstrated the value of CE-T1WI-based radiomics for preoperative evaluation of cell proliferation during the onset of meningioma, while Kim et al. (18) established a deep-learning model utilizing CE-T1WI for identification of the dural tail sign, significantly improving the clinical diagnostic performance for meningioma. Thus, the quantitative data derived from histogram analysis may provide deeper insights into histological characteristics. In this study, we explored the value of histogram analysis based on CE-T1WI in preoperatively distinguishing between gonadotroph and non-gonadotroph macroadenomas, as well as predicting the Ki-67 and P53 expression levels of pathological indicators. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1068/rc).

Methods

Study design and patients

There were 152 patients (64 males and 88 females; age range, 20–78 years; mean age: 49.6 years) with suspected pituitary macroadenomas treated at Henan Provincial People’s Hospital between January 2022 and September 2024 enrolled. Inclusion criteria included: (I) age ≥18 years; (II) preoperative MRI examination; (III) surgical resection and histopathological diagnosis completed within one week after MRI. Exclusion criteria included: (I) poor MRI image quality affecting diagnostic analysis; (II) large intratumoral hemorrhagic or cystic areas; (III) the pathological diagnosis was not a pituitary adenoma. Pathologists made the histopathological diagnosis based on the 2022 WHO classification of pituitary tumors (3). After that, 34 patients were excluded, out of which 10 had massive hemorrhage, 15 had cystic lesions within the tumor, and 9 were diagnosed as non-pituitary adenomas. Finally, 118 patients with pituitary macroadenomas (49 males and 69 females; range, 20–77 years; mean age: 50.1 years) were included for analysis in this study. Figure 1 shows the flowchart of patient enrollment.

Figure 1 Flowchart of the enrolled patients. MRI, magnetic resonance imaging.

MRI examination protocol

All patients underwent preoperative examination on a 3T MR scanner (Prisma; Siemens Healthineers, Erlangen, Germany). The scanning protocol included: axial T1-weighted imaging (T1WI) with repetition time (TR): 500 ms, echo time (TE): 12 ms, flip angle: 90°, scanning layers: 18, slice thickness: 2.0 mm, unit voxel: 0.6 mm × 0.6 mm × 2.0 mm, field of view (FOV): 230 mm × 230 mm; axial T2-weighted imaging (T2WI) with TR: 4,600 ms, TE: 97 ms, flip angle: 150°, scanning layers: 18, slice thickness: 2.0 mm, unit voxel: 0.6 mm × 0.6 mm × 2.0 mm, FOV: 230 mm × 230 mm; Coronal T1WI: TR: 500 ms, TE: 12 ms, flip angle: 120°, scanning layers: 11, slice thickness: 2.0 mm, unit voxel: 0.6 mm × 0.6 mm × 2.0 mm, FOV: 230 mm × 230 mm, scanning time: 91 seconds; coronal T2WI: TR: 2,000 ms, TE: 62 ms, flip angle: 120°, scanning layers: 11, slice thickness: 2.0 mm, unit voxel: 0.6 mm × 0.6 mm × 2.0 mm, FOV: 230 mm × 230 mm, scanning time: 88 seconds; CE-T1WI scanning was performed using Gd-DTPA (Bayer Schering Pharma AG, Berlin, Germany; /kg) administered intravenously at a bolus injection of 0.1 mmol/kg.

Imaging analysis

Two experienced radiologists (with 14 and 7 years of experience in MRI diagnosis of neurological neoplasms), blinded to the pathological results, independently analysed all imaging diagnosis, respectively. The raw MRI data were imported into 3D Slicer software (https://www.slicer.org), where regions of interest (ROIs) were manually delineated along the tumor margins on the largest coronal CE-T1WI slice, avoiding areas of necrosis, cystic degeneration, or hemorrhage. Then, the ROIs were copied to T1WI images to yield the subtraction between CE-T1WI and T1WI. Using SlicerRadiomics (https://www.radiomics.io/PyRadiomics), histogram parameters were extracted from the ROIs, including the 10th percentile (Perc.10%), 90th percentile (Perc.90%), kurtosis, mean, median, maximum, minimum, and skewness values, as shown in Figures 2,3. The histogram parameters were obtained from the CE-T1WI and the subtraction between CE-T1WI and T1WI, respectively.

Figure 2 A 43-year-old male with gonadotroph macroadenoma. (A) Coronal T1WI image; (B) coronal CE-T1WI image with mild enhancement of pituitary adenoma; (C) manual delineation of the tumor’s ROI on the CE-T1WI image; (D) histogram analysis was conducted on matched CE-T1WI and T1WI ROIs to enable comparative visualization of signal intensity distributions. Additionally, parameters derived from the subtraction between CE-T1WI and T1WI were compared with those derived from CE-T1WI; (E) the pathological examination of gonadotroph adenomas revealed a sparse arrangement of tumor cells with conspicuous mitotic figures (HE, ×400). CE-T1WI, contrast-enhanced T1-weighted imaging; HE, hematoxylin-eosin; ROI, region of interest; T1WI, T1-weighted imaging.

Figure 3 A 46-year-old male with somatotroph macroadenoma. (A) Coronal T1WI image; (B) coronal CE-T1WI image with mild enhancement of pituitary adenoma; (C) manual delineation of the tumor’s ROI on the CE-T1WI image; (D) histogram analysis was conducted on matched CE-T1WI and T1WI ROIs to enable comparative visualization of signal intensity distributions. Additionally, parameters derived from the subtraction between CE-T1WI and T1WI were compared with those derived from CE-T1WI; (E) the pathological examination of non-gonadotroph adenomas characterized by sheet-like growth of tumor cells, increasing cell density, and an elevating nuclear-cytoplasmic ratio (HE, ×400). CE-T1WI, contrast-enhanced T1-weighted imaging; HE, hematoxylin-eosin; ROI, region of interest; T1WI, T1-weighted imaging.

Knosp grading was applied as follows: grade 0, no extension beyond the medial carotid line; grade 1, pituitary adenomas extend beyond medial margins but not beyond the midpoint (intercarotid line) of the supra and intracavernous internal carotid artery (ICA); grade 2, pituitary adenomas extend beyond intercarotid line, but not beyond the lateral margins of the supra and intracavernous ICA; grade 3, the extension of pituitary adenomas exceed lateral margins of the supra and intracavernous ICA; and grade 4, pituitary adenomas completely encase the intracavernous ICA. Grades 3 and 4 are indicative of cavernous sinus invasion (12). Statistical analysis used parameter values derived from the mean of two radiologists’ measurements. Inter-observer repeatability of the measured parameter values was evaluated based on data measured by two radiologists.

Histopathological analysis

Surgical specimens of pituitary macroadenomas were subjected to histopathological analysis. Immunohistochemistry was employed to assess Ki-67 and P53 expression levels. Ki-67, a marker of cellular proliferation, was quantified by the percentage of positive cells, with high expression correlating with increased proliferative activity and poorer clinical outcomes. P53, a tumor suppressor, is frequently overexpressed or mutated in malignant tumors, indicating loss of genomic stability, and is associated with an adverse prognosis.

Statistical analysis

Statistical analysis was performed using SPSS 26.0 software (IBM SPSS Statistics; Chicago, IL, USA). Quantitative histogram parameters were expressed as mean ± standard deviation (x±s). Normality and homoscedasticity of the measurement data were assessed using the Shapiro-Wilk test and Levene test, respectively. For normally distributed data with equal variances, the independent samples t-test was utilized. If variances were unequal, Welch’s corrected t-test was employed. Non-normally distributed data were analyzed using the Mann-Whitney U test. Receiver operating characteristic (ROC) curves were plotted to assess the diagnostic performance of statistically significant parameters. The area under the curve (AUC), sensitivity, and specificity were determined based on optimal cutoff values derived from the ROC analysis. The Knosp grading system for pituitary adenomas was analyzed using the Pearson Chi-squared test to evaluate categorical correlations between tumor invasion grades and histopathological subtypes. For quantitative analysis, histogram parameters derived from CE-T1WI were correlated with Ki-67 and P53 expression levels using Pearson correlation coefficients. Statistical significance was defined as P<0.05.

Ethics

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The Ethics Committee of Henan Provincial People’s Hospital approved this retrospective study (No. 2021-148). Since the study utilized previously collected clinical data and maintained adherence to standard therapeutic protocols without intervention, obtaining informed consent was formally waived.

Results

Patient characteristics

The cohort comprised 118 patients, surgically resected pituitary macroadenomas with histopathologically confirmed diagnosis; 58 patients (49.2%) were classified as gonadotroph macroadenomas, while 60 patients (50.8%) were diagnosed as non-gonadotroph macroadenomas, including 15 somatotroph adenomas, 19 lactotroph adenomas, 18 corticotroph adenomas, and 8 null cell adenomas. The clinical data for all patients are summarized in Table 1. Non-gonadotroph macroadenomas showed a higher propensity for cavernous sinus invasion compared to gonadotroph macroadenomas (P=0.027). Among the 58 gonadotroph macroadenomas, 22 (37.9%) exhibited invasiveness (Knosp grade 4: 6 cases; Knosp grade 3: 16 cases). In contrast, 35 of the 60 non-gonadotroph macroadenomas (58.3%) showed invasiveness (Knosp grade 4: 15 cases; Knosp grade 3: 20 cases). Non-gonadotroph macroadenomas demonstrated significantly higher expression levels of Ki-67 and P53 pathological markers compared to gonadotroph macroadenomas (P=0.007 and P=0.042, respectively). Two radiologists performed blinded subtype classification independently with strong inter-observer agreement [intraclass correlation coefficient (ICC) =0.890]. Age showed no significant intergroup differences [gonadotroph macroadenomas (51.3±12.7 years) versus non-gonadotroph macroadenomas (48.9±14.2 years); t=1.39, P=0.169]. Sex distribution was comparable between groups (χ²=0.607, P=0.436).

Comparison of histogram parameters for differentiating gonadotroph from non-gonadotroph macroadenomas

All features showed excellent inter-observer agreement (ICC >0.850). Table 2 showed histogram parameters derived from CE-T1WI and the subtraction between CE-T1WI and T1WI, both performed significantly higher Perc.10%, Perc.90%, mean, median, maximum, and minimum values in non-gonadotroph macroadenomas compared to gonadotroph macroadenomas (P<0.001 for all). In addition, the parameters derived from the subtraction between CE-T1WI and T1WI demonstrated better discrimination performance compared to those derived from CE-T1WI. We employed the mean value as a reference metric to evaluate data stability, with normalization performed relative to frontal lobe signal intensity. The results demonstrated greater stability in the parameters derived from the subtraction between CE-T1WI and T1WI to white matter ratio (0.84±0.28) compared to those derived from CE-T1WI to white matter ratio (1.93±0.54), with statistically significant differentiation (P<0.001).

ROC analysis of histogram parameters in differentiating gonadotroph from non-gonadotroph macroadenomas

ROC curves were plotted for the histogram parameters (Perc.10%, Perc.90%, mean, median, maximum, and minimum values) to differentiate between gonadotroph and non-gonadotroph macroadenomas. All parameters were statistically significant (P<0.001 for all) showed in Table 3 and Figure 4. The histogram analysis derived from the subtraction between CE-T1WI and T1WI demonstrated better discrimination performance compared to that derived from CE-T1WI. For instance, Perc.10% derived from subtraction (AUC =0.979) had significantly greater AUC than that derived from CE-T1WI (AUC =0.838, P<0.05). The combined parameters achieved an AUC of 0.992, significantly outperforming individual parameters (P<0.05).

Figure 4 ROC curves of histogram parameters with statistical significance for distinguishing between gonadotroph macroadenomas and non-gonadotroph macroadenomas. (A) ROC curves of parameters derived from CE-T1WI; (B) ROC curves of parameters derived from the subtraction between CE-T1WI and T1WI. The AUC of the combined parameter is significantly higher than that of individual parameters, indicating superior overall diagnostic performance. AUC, area under the curve; CE-T1WI, contrast-enhanced T1-weighted imaging; Perc.10%, the 10th percentile; Perc.90%, the 90th percentile; ROC, receiver operating characteristic; T1WI, T1-weighted imaging.

Correlation between CE-T1WI histogram parameters and Ki-67/P53 expression levels

According to the 2018 European Society of Endocrinology clinical practice guidelines for aggressive pituitary tumors (19), Ki-67 and P53 expression levels are closely associated with tumor aggressiveness. P53 positive rates and high Ki-67 indices (≥3%) indicate higher invasiveness. In Table 2, postoperative pathological analysis revealed that Ki-67 and P53 expression levels were significantly higher in non-gonadotroph macroadenomas than in gonadotroph macroadenomas [Ki-67, 5.00 (3.00, 6.00) versus 4.00 (3.00, 5.00), P=0.007; P53, 5.00 (4.00, 6.00) versus 4.00 (3.00, 6.00), P=0.042]. Table 4 showed that CE-T1WI histogram parameters such as maximum, minimum, and Perc.90% held positive correlations with Ki-67 expression levels (r=0.259, 0.208, and 0.211, respectively; P<0.05 for all), with the maximum exhibiting the strongest correlation, as shown in Figure 5A-5C. Additionally, the minimum, Perc.10% and Perc.90% values held positive correlations with P53 expression levels (r=0.201, 0.193 and 0.244, respectively; P<0.05 for all), with the Perc.90% exhibiting the strongest correlation, as shown in Figure 5D-5F.

Figure 5 Scatter plots show significant correlations between CE-T1WI histogram parameters and Ki-67/P53 expression levels. The maximum values exhibit the strongest positive correlation with Ki-67 indices (r=0.259, P=0.005), and the Perc.90% values exhibit the strongest positive correlation with P53 protein (r=0.244, P=0.008). CE-T1WI, contrast-enhanced T1-weighted imaging; Perc.10%, the 10th percentile; Perc.90%, the 90th percentile.

Discussion

This study utilized CE-T1WI histogram analysis to differentiate between gonadotroph and non-gonadotroph macroadenomas. The results demonstrated that non-gonadotroph macroadenomas exhibited significantly higher values for Perc.10%, Perc.90%, mean, median, maximum, and minimum compared to gonadotroph macroadenomas, and the combined parameter model showed superior diagnostic performance. The parameters derived from the subtraction between CE-T1WI and T1WI exhibited superior diagnostic performance and stability compared to those derived from CE-T1WI. Non-gonadotroph macroadenomas were also more likely to invade the cavernous sinus than gonadotroph macroadenomas, with significantly higher Ki-67/P53 expression levels. Histogram parameters demonstrated preoperative predictive value for Ki-67/P53 expression levels, with maximum, minimum, and Perc.90% exhibiting positive correlations with Ki-67 expression levels, and the minimum, Perc.10% and Perc.90% values holding positive correlations with P53 expression levels.

Conventional MRI is inadequate in quantifying the biological characteristics of tumors, dynamically evaluating tumor behavior, and sensitively detecting intricate pathophysiological changes, thereby posing substantial challenges in distinguishing between gonadotroph and non-gonadotroph macroadenomas (9). CE-T1WI serves as a routine scanning protocol, accurately reflecting the anatomical structure, vascular supply, and cellular composition of lesions while thoroughly elucidating their internal characteristics (20). Histogram analysis quantifies the internal consistency of tumors by leveraging image data from the ROI to unveil features that are elusive to conventional imaging. This study explored the value of CE-T1WI histogram analysis in differentiating these two subtypes, with eight histogram parameters derived from ROI, six of which (Perc.10%, Perc.90%, maximum, mean, median, and minimum) showed statistical significance. The combined parameter model constructed from these six parameters exhibits superior diagnostic efficacy. The significant parameters in non-gonadotroph macroadenomas may reflect greater tumor blood supply and vascular permeability, as pituitary adenoma enhancement indirectly correlates with these factors. Their alterations in the structure and function of newly formed vessels damage the vascular basement membrane, increasing microvascular permeability and facilitating Gd-DTPA entry into tumor tissue, thereby enhancing invasiveness on imaging (21,22). Increased invasiveness is further linked to elevated cell proliferation rates, driving tumor expansion and neovascularization (15). In addition to the above factors, the T1WI signal intensities of adenomas exhibit interpatient variability due to other confounding factors, including age-related changes, edema, and fibrotic components. The subtraction methodology, by utilizing individual baseline T1WI values, CE-T1WI values, and computing differential signal alterations, effectively eliminates inherent adenoma tissue signals while preserving true enhancement patterns (23), thereby enhancing parametric comparability. This study revealed that the histogram analysis derived from the subtraction between CE-T1WI and T1WI demonstrated better discrimination performance compared to that derived from CE-T1WI. The subtraction-derived parameters could mitigate measurement biases due to adenoma signal heterogeneity, including lipid signal variations and magnetic field inhomogeneities, yielding more reliable enhancement quantification with markedly improved parametric stability.

The Knosp grading system is widely used for the preoperative assessment of pituitary adenomas (12). A study has shown that Knosp grades 3 and 4 often indicate cavernous sinus invasion, while grades 1 or 2 typically indicate no invasion (10). This study explored the differences in invasiveness between the two subtypes. The results showed that non-gonadotroph macroadenomas exhibited significantly higher cavernous sinus invasion possibilities (Knosp grades 3 and 4: 58.3% versus 37.9%, P=0.027), along with elevated Ki-67 (P=0.007) and P53 (P=0.042) expression levels compared to gonadotroph macroadenomas. Raverot et al. (15) reported that non-gonadotroph macroadenomas, particularly lactotroph adenomas and corticotroph adenomas, demonstrate higher rates of malignancy and vascular invasion, whereas gonadotroph macroadenomas exhibit a higher degree of cellular differentiation and relatively lower vascular invasion. Previous studies have also confirmed that non-gonadotroph pituitary adenomas are more aggressive (4,24), consistent with McCormack et al.’s findings that gonadotroph adenomas are less common among invasive pituitary tumors and pituitary carcinomas (24). This may be attributed to the lower nuclear-to-cytoplasmic ratio of gonadotroph adenoma cells (25,26), which exhibit well-differentiated cells and lower proliferative activity (27,28).

Ki-67 is a nuclear protein intricately linked to cellular proliferation, frequently utilized as an indicator of tumor progression and recurrence, serving as a pivotal marker for evaluating tumor invasiveness (29). Previous studies (30,31) found that Ki-67 expression levels correlate positively with tumor size and invasiveness, indicating the relationship between Ki-67 expression levels and pituitary adenoma aggressiveness. This study demonstrated that non-gonadotroph macroadenomas exhibited significantly higher Ki-67 (P=0.007) and P53 (P=0.042) expression levels compared to gonadotroph macroadenomas. Histogram parameters significantly correlated with Ki-67 expression levels, and three parameters (maximum, minimum, and Perc.90%) were identified with predictive value of Ki-67 expression levels, all of which were positively correlated with Ki-67 (P<0.05 for all); the maximum showed the strongest correlation (r=0.259, P<0.01). This may be attributed to the fact that high Ki-67 indices indicate greater invasiveness, which is associated with richer blood supply, more significant enhancement, and increased vascular permeability due to tissue destruction (32). However, the mean showed no significant correlation with Ki-67 expression levels (P=0.129), possibly due to the high cell density associated with elevated Ki-67 expression in invasive pituitary adenomas, resulting in reduced intercellular spaces. Additionally, the rapid proliferation of highly invasive tumor cells leads to uneven cell division, rendering the tumor more susceptible to cystic degeneration, hemorrhage, and necrosis (32), thereby impeding contrast agent penetration. Based on postoperative pathological reports, this study revealed significantly higher invasiveness in non-gonadotroph macroadenomas. This finding aligns with the Knosp grading results and accords with previous studies (4), which reported that non-gonadotroph macroadenomas are more likely to invade the cavernous sinus than gonadotroph macroadenomas (55% versus 26%, P=0.026). Conversely, Ugga et al. reported that gonadotroph adenomas often lack Ki-67 expression, indicating limited proliferative capacity and reduced invasiveness (33,34).

P53 protein serves as a critical tumor suppressor and key regulator of tumor proliferation and invasion (13). In this study, positive correlations were observed between P53 expression levels and histogram parameters, including the minimum, Perc.10%, and Perc.90% values (P<0.05 for all), with the Perc.90% showing the strongest correlation (r=0.244, P<0.01). The Perc.90%, reflecting hyperintensity of tumor regions excluding outliers, such as vascular or necrotic areas, potentially mediated by increased invasiveness, micro-necrotic proliferation, and microstructural heterogeneity (19). Additionally, enhanced neovascularization and vascular permeability driven by P53 overexpression may further amplify Gd-DTPA uptake, augmenting signal intensity. Similarly, the association of minimum and Perc.10% values with P53 levels likely reflects P53-induced microstructural complexity, hypervascularity, and cellular heterogeneity (35), collectively shaping tumor imaging phenotypes. Research on refractory pituitary adenomas and pituitary carcinomas highlights that invasive adenomas are characterized by elevated P53 levels, rapid growth, frequent recurrence, and resistance to conventional therapies (36), suggesting a positive correlation between high biomarker expression levels and tumor aggressiveness.

This study offers insights into the role of CE-T1WI and the subtraction between CE-T1WI and T1WI histogram analysis in differentiating gonadotroph from non-gonadotroph macroadenomas. However, there are several limitations. First, this study is a retrospective study with a relatively small sample size, which may induce substantial selection bias. Future prospective studies with larger, balanced cohorts and comprehensive assessments are recommended to substantiate these findings. Second, the manual delineation of ROIs, while carefully excluding hemorrhagic or cystic areas, induces significant selection bias that may impact the validity. Artificial intelligence (AI)-driven automated segmentation methods are needed to improve this bias. Third, this study was conducted at a single center, potentially limiting the generalizability of the results. Multi-center validation is necessary to extend the generality of the histogram model. Besides, this analysis exclusively examines differentiation based on CE-T1WI and does not incorporate other advanced imaging sequences for a combined analysis, which highlights the need for future comprehensive studies that utilize new imaging technologies or sequences to provide a more holistic understanding of pituitary adenomas and their invasiveness.

Conclusions

In summary, histogram analysis of both CE-T1WI and the subtraction between CE-T1WI and T1WI could differentiate gonadotroph macroadenomas from non-gonadotroph macroadenomas. Additionally, the CE-T1WI histogram parameters were correlated with the Ki-67 and P53 expression levels.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1068/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1068/dss

Funding: This study was supported by the National Natural Science Foundation of China (Nos. 82441022 and 82471963), and Open Project of Key Laboratory of Science and Engineering for the Multi-modal Prevention and Control of Major Chronic Diseases Ministry of Industry and Information Technology (No. MCD-2023-1-17).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1068/coif). M.W. reports that this study was supported by the National Natural Science Foundation of China (Nos. 82441022 and 82471963), and Open Project of Key Laboratory of Science and Engineering for the Multi-modal Prevention and Control of Major Chronic Diseases Ministry of Industry and Information Technology (No. MCD-2023-1-17). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The Ethics Committee of Henan Provincial People’s Hospital approved this retrospective study (No. 2021-148). Since the study utilized previously collected clinical data and maintained adherence to standard therapeutic protocols without intervention, obtaining informed consent was formally waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ostrom QT, Gittleman H, Liao P, Vecchione-Koval T, Wolinsky Y, Kruchko C, Barnholtz-Sloan JS. CBTRUS Statistical Report: Primary brain and other central nervous system tumors diagnosed in the United States in 2010-2014. Neuro Oncol 2017;19:v1-v88. [Crossref] [PubMed]
2. Molitch ME. Diagnosis and Treatment of Pituitary Adenomas: A Review. JAMA 2017;317:516-24. [Crossref] [PubMed]
3. Asa SL, Mete O, Perry A, Osamura RY. Overview of the 2022 WHO Classification of Pituitary Tumors. Endocr Pathol 2022;33:6-26. [Crossref] [PubMed]
4. Bai Y, Shen Y, Chen R, Zhang X, Nittka M, Koerzdoerfer G, Gong Q, Wang M. Magnetic resonance fingerprinting for preoperative differentiation between gonadotroph and non-gonadotroph pituitary macroadenomas. Eur Radiol 2021;31:8420-8. [Crossref] [PubMed]
5. Øystese KA, Casar-Borota O, Normann KR, Zucknick M, Berg JP, Bollerslev J. Estrogen Receptor α, a Sex-Dependent Predictor of Aggressiveness in Nonfunctioning Pituitary Adenomas: SSTR and Sex Hormone Receptor Distribution in NFPA. J Clin Endocrinol Metab 2017;102:3581-90. [Crossref] [PubMed]
6. Lee M, Lupp A, Mendoza N, Martin N, Beschorner R, Honegger J, Schlegel J, Shively T, Pulz E, Schulz S, Roncaroli F, Pellegata NS. SSTR3 is a putative target for the medical treatment of gonadotroph adenomas of the pituitary. Endocr Relat Cancer 2015;22:111-9. [Crossref] [PubMed]
7. Feng Z, He D, Mao Z, Wang Z, Zhu Y, Zhang X, Wang H. Utility of 11C-Methionine and 18F-FDG PET/CT in Patients With Functioning Pituitary Adenomas. Clin Nucl Med 2016;41:e130-4. [Crossref] [PubMed]
8. Sumida M, Uozumi T, Mukada K, Arita K, Kurisu K, Yano T, Onda J, Satoh H, Ikawa F. MRI of pituitary adenomas: the position of the normal pituitary gland. Neuroradiology 1994;36:295-7. [Crossref] [PubMed]
9. Hagiwara A, Inoue Y, Wakasa K, Haba T, Tashiro T, Miyamoto T. Comparison of growth hormone-producing and non-growth hormone-producing pituitary adenomas: imaging characteristics and pathologic correlation. Radiology 2003;228:533-8. [Crossref] [PubMed]
10. Davies BM, Carr E, Soh C, Gnanalingham KK. Assessing size of pituitary adenomas: a comparison of qualitative and quantitative methods on MR. Acta Neurochir (Wien) 2016;158:677-83. [Crossref] [PubMed]
11. Xue C, Liu S, Deng J, Liu X, Li S, Zhang P, Zhou J. Apparent Diffusion Coefficient Histogram Analysis for the Preoperative Evaluation of Ki-67 Expression in Pituitary Macroadenoma. Clin Neuroradiol 2022;32:269-76. [Crossref] [PubMed]
12. Knosp E, Steiner E, Kitz K, Matula C. Pituitary adenomas with invasion of the cavernous sinus space: a magnetic resonance imaging classification compared with surgical findings. Neurosurgery 1993;33:610-7; discussion 617-8. [Crossref] [PubMed]
13. Hasanov R, Aydoğan Bİ, Kiremitçi S, Erden E, Güllü S. The Prognostic Roles of the Ki-67 Proliferation Index, P53 Expression, Mitotic Index, and Radiological Tumor Invasion in Pituitary Adenomas. Endocr Pathol 2019;30:49-55. [Crossref] [PubMed]
14. Chiloiro S, Bianchi A, Doglietto F, de Waure C, Giampietro A, Fusco A, Iacovazzo D, Tartaglione L, Di Nardo F, Signorelli F, Lauriola L, Anile C, Rindi G, Maira G, Pontecorvi A, De Marinis L. Radically resected pituitary adenomas: prognostic role of Ki 67 labeling index in a monocentric retrospective series and literature review. Pituitary 2014;17:267-76. [Crossref] [PubMed]
15. Raverot G, Ilie MD, Lasolle H, Amodru V, Trouillas J, Castinetti F, Brue T. Aggressive pituitary tumours and pituitary carcinomas. Nat Rev Endocrinol 2021;17:671-84. [Crossref] [PubMed]
16. Xue C, Zhou Q, Zhang P, Zhang B, Sun Q, Li S, Deng J, Liu X, Zhou J. MRI histogram analysis of tumor-infiltrating CD8+ T cell levels in patients with glioblastoma. Neuroimage Clin 2023;37:103353. [Crossref] [PubMed]
17. Ouyang ZQ, He SN, Zeng YZ, Zhu Y, Ling BB, Sun XJ, Gu HY, He B, Han D, Lu Y. Contrast enhanced magnetic resonance imaging-based radiomics nomogram for preoperatively predicting expression status of Ki-67 in meningioma: a two-center study. Quant Imaging Med Surg 2023;13:1100-14. [Crossref] [PubMed]
18. Kim H, Kim HG, Oh JH, Lee KM. Deep-learning model for diagnostic clue: detecting the dural tail sign for meningiomas on contrast-enhanced T1 weighted images. Quant Imaging Med Surg 2023;13:8132-43. [Crossref] [PubMed]
19. Raverot G, Burman P, McCormack A, Heaney A, Petersenn S, Popovic V, Trouillas J, Dekkers OMEuropean Society of Endocrinology. European Society of Endocrinology Clinical Practice Guidelines for the management of aggressive pituitary tumours and carcinomas. Eur J Endocrinol 2018;178:G1-G24. [Crossref] [PubMed]
20. Kandemirli SG, Chopra S, Priya S, Ward C, Locke T, Soni N, Srivastava S, Jones K, Bathla G. Presurgical detection of brain invasion status in meningiomas based on first-order histogram based texture analysis of contrast enhanced imaging. Clin Neurol Neurosurg 2020;198:106205. [Crossref] [PubMed]
21. Yeo DM, Oh SN, Jung CK, Lee MA, Oh ST, Rha SE, Jung SE, Byun JY, Gall P, Son Y. Correlation of dynamic contrast-enhanced MRI perfusion parameters with angiogenesis and biologic aggressiveness of rectal cancer: Preliminary results. J Magn Reson Imaging 2015;41:474-80. [Crossref] [PubMed]
22. Chen YL, Li R, Chen TW, Ou J, Zhang XM, Chen F, Wu L, Jiang Y, Laws M, Shah K, Joseph B, Hu J. Whole-tumour histogram analysis of pharmacokinetic parameters from dynamic contrast-enhanced MRI in resectable oesophageal squamous cell carcinoma can predict T-stage and regional lymph node metastasis. Eur J Radiol 2019;112:112-20. [Crossref] [PubMed]
23. Ma YJ, Shao H, Fan S, Lu X, Du J, Young IR, Bydder GM. New options for increasing the sensitivity, specificity and scope of synergistic contrast magnetic resonance imaging (scMRI) using Multiplied, Added, Subtracted and/or FiTted (MASTIR) pulse sequences. Quant Imaging Med Surg 2020;10:2030-65. [Crossref] [PubMed]
24. McCormack A, Dekkers OM, Petersenn S, Popovic V, Trouillas J, Raverot G, Burman P. ESE survey collaborators. Treatment of aggressive pituitary tumours and carcinomas: results of a European Society of Endocrinology (ESE) survey 2016. Eur J Endocrinol 2018;178:265-76. [Crossref] [PubMed]
25. Saccomanno K, Bassetti M, Lania A, Losa M, Faglia G, Spada A. Immunodetection of glycoprotein hormone subunits in nonfunctioning and glycoprotein hormone-secreting pituitary adenomas. J Endocrinol Invest 1997;20:59-64. [Crossref] [PubMed]
26. Obad-Kovačević D, Kardum-Skelin I, Jelić-Puškarić B, Vidjak V, Blašković D. Parotid gland tumors: correlation between routine cytology and cytomorphometry by digital image analysis using conventional and newly introduced cytomorphometric parameters. Diagn Cytopathol 2013;41:776-84. [Crossref] [PubMed]
27. Almeida JP, Stephens CC, Eschbacher JM, Felicella MM, Yuen KCJ, White WL, Mooney MA, Bernat AL, Mete O, Zadeh G, Gentili F, Little AS. Clinical, pathologic, and imaging characteristics of pituitary null cell adenomas as defined according to the 2017 World Health Organization criteria: a case series from two pituitary centers. Pituitary 2019;22:514-9. [Crossref] [PubMed]
28. Chaidarun SS, Klibanski A. Gonadotropinomas. Semin Reprod Med 2002;20:339-48. [Crossref] [PubMed]
29. Zhang B, Zhou F, Zhou Q, Xue C, Ke X, Zhang P, Han T, Deng L, Jing M, Zhou J. Whole-tumor histogram analysis of multi-parametric MRI for differentiating brain metastases histological subtypes in lung cancers: relationship with the Ki-67 proliferation index. Neurosurg Rev 2023;46:218. [Crossref] [PubMed]
30. Nishioka H, Haraoka J, Akada K. Growth potential of prolactinomas in men: is it really different from women? Surg Neurol 2003;59:386-90; discussion 390-1. [Crossref] [PubMed]
31. Mete O, Ezzat S, Asa SL. Biomarkers of aggressive pituitary adenomas. J Mol Endocrinol 2012;49:R69-78. [Crossref] [PubMed]
32. Yuan J, Xie D, Fang S, Meng F, Wu Y, Shan D, Shao N, Wang B, Tian Z, Wang Y, Xu C, Chen X. Qualitative and quantitative MRI analysis of alveolar soft part sarcoma: correlation with histological grade and Ki-67 expression. Insights Imaging 2024;15:142. [Crossref] [PubMed]
33. Ugga L, Cuocolo R, Solari D, Guadagno E, D’Amico A, Somma T, Cappabianca P, Del Basso de Caro ML, Cavallo LM, Brunetti A. Prediction of high proliferative index in pituitary macroadenomas using MRI-based radiomics and machine learning. Neuroradiology 2019;61:1365-73. [Crossref] [PubMed]
34. Schreiber S, Saeger W, Lüdecke DK. Proliferation markers in different types of clinically non-secreting pituitary adenomas. Pituitary 1999;1:213-20. [Crossref] [PubMed]
35. Lin L, Bhawana R, Xue Y, Duan Q, Jiang R, Chen H, Chen X, Sun B, Lin H. Comparative Analysis of Diffusional Kurtosis Imaging, Diffusion Tensor Imaging, and Diffusion-Weighted Imaging in Grading and Assessing Cellular Proliferation of Meningiomas. AJNR Am J Neuroradiol 2018;39:1032-8. [Crossref] [PubMed]
36. Dai C, Liu X, Ma W, Wang R. The Treatment of Refractory Pituitary Adenomas. Front Endocrinol (Lausanne) 2019;10:334. [Crossref] [PubMed]