Pineal region tumors: prognostic stratification based on magnetic resonance imaging features—a retrospective cohort study

Introduction

The pineal region, located in the central midline of the brain, encompasses several vital structures, including the pineal gland, the quadrigeminal cistern, the cistern of the velum interpositum, the posterior part of the third ventricle, and the surrounding tissue around the brainstem, thalamus, and corpus callosum (1). This region consists of a complex composition of glial cells, large pineal cells, immature small pineal cells, and various vascular structures. Pineal region tumors are characterized by diverse pathologies and cover a wide spectrum of subtypes, including germ cell tumors, pineal parenchymal cell tumors, meningiomas, gliomas, and more (2).

Pineal region tumors are relatively rare among intracranial tumors, accounting for approximately 1–3% of all intracranial tumors, with a slightly higher prevalence of 4% in children (3) and less than 1% in adults. Among these, germ cell tumors and pineal parenchymal tumors are the most common (4). Germ cell tumors include several subtypes, such as germinomas, embryonal carcinomas, yolk sac tumors, choriocarcinomas, teratomas, and mixed germ cell tumors. Pineal parenchymal tumors, on the other hand, encompass pineocytomas (PC), pineal parenchymal tumors of intermediate differentiation (PPTID), papillary tumors of the pineal region (PTPR), pinealoblastoma (PB), and desmoplastic myxoid tumors of the pineal region with SMARCB1 mutations (5).

Most previous studies have focused on identifying various types of tumors in the pineal region. For example, the apparent diffusion coefficient (ADC) value of germinoma is generally higher than that of PC (6). Researchers from Tiantan Hospital developed a diagnostic model to differentiate between pineal germinoma and pineoblastoma, achieving a diagnostic accuracy of 94% in the validation set (7). However, we found no studies that integrated tumor prognosis with imaging findings. Our study proposes that specific imaging characteristics may serve as prognostic predictors. Pineal region tumors often exhibit varying prognoses due to their distinct pathological characteristics, and their imaging features can differ significantly. For instance, while germinomas are malignant, they typically have a good prognosis due to their sensitivity to radiotherapy, as noted in a recent clinical trial (8). Consequently, preliminary stratification of prognosis based on preoperative imaging features of pineal region tumors could enhance the precision of clinical treatment selection and standardize patient management protocols. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-24-281/rc).

Methods

Patients

This is a retrospective cohort study. Between January 2010 and June 2023, a total of 33 patients diagnosed with pineal region tumors were treated at the Seventh Medical Center of the Chinese PLA General Hospital. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by institutional ethics committee board of the Seventh Medical Center of Chinese PLA General Hospital (No. S2024-041-01) and individual consent for this retrospective analysis was waived.

Clinical characteristics

For each of these 33 patients, we collected comprehensive clinical data, including initial and presenting symptoms, age, sex, and pathology results. The inclusion criteria were as follows: first, patients underwent preoperative or pre-treatment magnetic resonance imaging (MRI) or computed tomography (CT) scans. Second, each patient had either post-surgical or post-biopsy pathology results. Third, the pathological diagnosis indicated a neoplasm of pineal origin or located within the pineal region.

Imaging protocols

All patients underwent preoperative brain T2-weighted imaging (T2WI), T1-weighted FLAIR imaging (T1WI-FLAIR), diffusion-weighted imaging (DWI), and contrast-enhanced T1-weighted imaging (CE-T1WI). Susceptibility-weighted imaging (SWI) was performed in some patients. Patients were required to remove metal objects and provide informed consent before the examination. For children who could not cooperate, oral chloral hydrate was considered before the examination. MRI images were acquired using either a 3T or 1.5T scanner with the 8HRBRAIN radiofrequency coil (GE Medical Systems and Siemens Medical Systems). The acquisition parameters for the T2WI sequence were as follows: repetition time (TR) =5,000 ms, echo time (TE) =119 ms, slice thickness =6 mm, slice spacing =7 mm, pixel spacing =1 mm × 1 mm, field of view (FOV) =220 mm × 220 mm, flip angle =90°. The acquisition parameters for the T1WI sequence were as follows: TR =1,964 ms, TE =29 ms, slice thickness =6 mm, slice spacing =7 mm, pixel spacing =1 mm × 1 mm, FOV =220 mm × 220 mm, flip angle =90°. CE-T1WI was performed using the T1WI sequence parameters after the rapid injection of a gadolinium-gadopentetate dimeglumine (DTPA) contrast agent (0.1 mmol/kg Gadovist). DWI was conducted with two different motion-probing gradient values (b=0 and 1,000 s/mm²) applied in three orthogonal directions. The acquired images were combined to form a single composite DWI, and their ADC values were calculated for each pixel to generate ADC images.

Imaging evaluation methods

The evaluation of pineal region tumors was conducted in accordance with conventional radiological diagnostic criteria. This comprehensive evaluation encompassed eight key aspects, as follows:

• Maximum diameter of mass: the tumor’s maximum diameter was measured as its length.
• Morphology of the lesion: lesions were classified as regular or irregular. A lesion was considered regular if it appeared superficially lobulated, rounded, or had a regular wedged shape. Conversely, deeply lobulated or “gaping” lesions were categorized as irregular.
• Borders of the lesion: on T2WI sequences, lesions were considered well-defined if they exhibited clear demarcation from surrounding tissues; otherwise, they were classified as poorly defined.
• Intensity: the intensity of the lesion on T2WI and T1WI was classified as homogeneous or heterogeneous.
• Intensity on DWI: the intensity on DWI was classified as homogeneous hyperintense, homogeneous iso-intense, or heterogeneous.
• Enhancement: the enhancement of lesions on post-contrast images was categorized as homogeneous enhancement, heterogeneous enhancement, or no enhancement.
• Intratumoral hemorrhage: in this study, intratumoral hemorrhage was defined as demonstrating hyperintensity on T1-weighted images and hypointensity on T2-weighted images (9). The presence of patchy hypointensity on SWI did not exclude a lesion from being classified as intratumoral hemorrhage, according to the study’s criteria. Notably, microbleeds observed on SWI within the lesion were differentiated and not considered part of the hemorrhage category for the purposes of this investigation.
• Cystic change: lesions with cysts were evaluated based on T2WI images. A cystic lesion was characterized by hyperintensity on T2WI and hypointensity on T1WI, with a well-defined border and regular, round-like morphology. Notably, patches of poorly defined, bordering hyperintensity on T2WI were not considered cystic changes.

A detailed illustration is shown in Figures 1-3. A radiologist with 3 years of experience in radiographic diagnosis was responsible for assessing these imaging features. Subsequently, another radiologist with 8 years of experience independently re-evaluated the images. If discrepancies arose in the assessment, the results of the senior doctor were used as the standard. The Kappa value and ICC value were used to calculate the consistency between the two assessments.

Figure 1 The morphology, border characteristics, and intensity of the pineal region tumors on T2WI and T1WI. The white arrows indicate the location of the lesions. T2WI, T2-weighted imaging; T1WI, T1-weighted imaging.

Figure 2 The varying intensities on DWI and the degree or patterns of enhancement of the pineal region tumors. The white arrows indicate the location of the lesions. DWI, diffusion-weighted imaging.

Figure 3 The presence of intratumoral hemorrhage and cystic changes within the pineal region tumors. The white arrows indicate the location of the lesions. T2WI, T2-weighted imaging; T1WI, T1-weighted imaging; T1WI + C, contrast-enhanced T1-weighted imaging.

The standard of prognostic stratification

To establish the prognostic stratification standard, we conducted an extensive review of more than 50 papers focusing on pineal region tumors. We collected and collated data on the survival rates and prognoses associated with each tumor type, which are presented in Table 1.

The grading criteria were established as follows: an overall survival (OS) rate of ≥90% was categorized as a good prognosis, while an OS rate of <90% was categorized as a poor prognosis.

Tumors without an overall survival rate in the table were graded based on a comprehensive analysis of World Health Organization (WHO) grading, median survival time, 3-year survival rate, and other relevant factors.

Statistical analysis

The statistical software SPSS (version 26.0) was used to perform the statistical analysis. The Fisher exact probability method, one-way analysis of variance (ANOVA) test, and binary logistic regression were applied in this study. A two-sided P value of <0.05 was considered statistically significant (19). Normally distributed quantitative data were presented as mean ± standard deviation (SD), and categorical data were expressed as frequency (percentage). Inter-observer agreement was assessed using Cohen’s Kappa statistic, with the following interpretation: Kappa <0.2 indicates poor agreement; Kappa between 0.21 and 0.4 indicates fair agreement; Kappa between 0.41 and 0.6 indicates moderate agreement; Kappa between 0.61 and 0.8 indicates strong agreement; Kappa between 0.81 and 1.0 indicates excellent agreement.

Results

The baseline characteristics and clinical characteristics are shown in Table 2. Thirty-three patients (6 females and 27 males) were consecutively enrolled in this study. The flow diagram of the study population is shown in Figure 4. The age range of the patients was from 1 to 72 years, with a mean age of 26.4 years (SD =17.7 years). Pineal region tumors often lack distinctive clinical characteristics. The most common initial presentation was intracranial hypertension, primarily due to obstructive hydrocephalus. Other symptoms included headache, dizziness, blurred vision, unsteady gait, polydipsia, and abnormal development of sexual characteristics. Among the 33 patients, 25 initially experienced headaches and dizziness, while 23 presented with symptoms of obstructive hydrocephalus. Additionally, 5 patients reported nausea and vomiting, and 3 had complaints of decreased or blurred vision.

Figure 4 Flow diagram of the study population. PLA, People’s Liberation Army of China; MR, magnetic resonance.

Pathologic diagnosis

Among the 33 patients, all underwent either biopsy or surgical resection, resulting in clear postoperative pathology results.

The distribution of tumor types among the 33 cases was as follows: germinomas accounted for 27.3%, mixed germ cell tumors for 18.2%, epidermoid cysts for 12.1%, PPTID for 9.1%, PTPR for 6.1%, immature teratoma for 6.1%, and the following tumor types each accounted for 3.0%: mature teratoid tumor, adult-type diffuse gliomas (astrocytoma, IDH-mutant), meningiomas (WHO grade 1), atypical meningiomas (WHO grade 2), supratentorial ependymomas, atypical teratoid/rhabdoid tumors (AT/RT), and glioblastomas (IDH-wildtype), as shown in Table 3.

As shown in Table 4, a one-way ANOVA test was used to determine the independent clinical risk features for prognosis. We found that only intratumoral hemorrhage (P=0.017) significantly influenced tumor prognosis. Although the enhancement pattern and T1-weighted image features were not statistically significant in the one-way ANOVA test, they became significant when the sample size was slightly larger, so they were included in the logistic regression analysis. Additionally, while age did not show a statistically significant difference between prognostic groups, it has an important clinical impact on prognosis. Therefore, multivariate statistical analysis was conducted on intratumoral hemorrhage, enhancement patterns, T1-weighted image features, and patient age using binary logistic regression. As shown in Table 5, intratumoral hemorrhage (P=0.015, OR =18.998, 95% CI: 1.780–202.798) was identified as a significant factor affecting the prognosis of pineal tumors. Specifically, the presence of intratumoral hemorrhage increased the likelihood of a poor prognosis by nearly 19.0 times compared to tumors without intratumoral hemorrhage. Inter-observer agreement is also shown in Table 4.

Discussion

The pineal region comprises a diverse array of components, including 13 histological types of tumors in our study. These tumors exhibit a spectrum of prognoses, underscoring the importance of preoperative prognostic stratification in guiding clinical management. Previous studies on pineal region tumors have typically involved small cohorts (6,20), with the largest including 130 patients (7), whereas our study, which involved 33 patients, provides a more representative dataset.

To the best of our knowledge, the existing literature lacks extensive exploration of the correlation between imaging features of pineal region tumors and their prognostic implications. At the same time, MRI has proven effective in distinguishing between multiple intracranial tumor types, with high accuracy (21). In this study, we used postoperative pathology as the gold standard, integrating it with prognostic data from previous literature to classify tumors into two prognostic categories based on overall survival. Subsequently, we evaluated the imaging features of each patient and conducted statistical analyses to identify imaging characteristics suggestive of prognosis in pineal region tumors. MRI scans are essential for delineating the morphology, location, margins, and histological origin of pineal region tumors. In our analysis of 33 cases, germinomas emerged as the predominant subtype, constituting 27.3% of cases, consistent with previous reports (1,2,4). Notably, germinomas accounted for 39.1% of germ cell tumors in our cohort, with a significant male predominance, aligning with historical observations (1), with a male-to-female ratio ranging from 2 to 17 times (2).

This study utilized preoperative MRI to highlight the adverse impact of intratumoral hemorrhage on the prognosis of pineal region tumors. We found that tumors with intratumoral hemorrhage were nearly 19.0 times more likely to have a poor prognosis compared to those without it. Among tumors with poor prognosis, pineal parenchymal tumors (22), mixed germ cell tumors, and immature teratomas frequently exhibit intratumoral hemorrhage, as illustrated in Figure 5. In contrast, intratumoral hemorrhage was less common in germinomas (3 out of 9 cases), which generally indicates a good prognosis.

Figure 5 These were a diagram of intratumoral hemorrhage in some tumors with a poor prognosis. (A,B) intratumoral hemorrhage of an atypical meningioma; (C,D) intratumoral hemorrhage of a glioblastoma; (E,F) intratumoral hemorrhage of a mixed germ cell tumor. The white arrows indicate the location of intratumoral hemorrhage. T2WI, T2-weighted imaging; T1WI, T1-weighted imaging.

A previous study reported microbleeds within germinomas of the basal ganglia, which typically appeared as small hypointense dots on SWI (23). This finding is distinct from the flaky pattern indicative of intratumoral hemorrhage identified in this study. The discrepancy in observations may arise from the different criteria and features used to define intratumoral hemorrhage. Moreover, due to the limited number of patients who underwent SWI in this study, a subset was evaluated using T1WI and T2WI for the presence or absence of intratumoral hemorrhage. Consequently, the definition of intratumoral hemorrhage in this study did not align with the conventional definition of microbleeds observed on SWI. Future studies will aim to enroll a larger cohort scanned with SWI to discern the prognostic significance of patchy hemorrhage and micro-hemorrhage in pineal region tumors.

In this study, the enhancement patterns did not reach statistical significance, which may be attributed to the insufficient sample size. However, a trend towards significance was observed, suggesting that the role of enhancement patterns is not negligible. When patients without pathological confirmation but with clinical observations suggestive of germinomas were included in the analysis, the enhancement patterns became statistically significant, as shown in Table 6. Therefore, the study highlighted that tumors demonstrating homogeneous enhancement on MRI tend to be associated with a reduced likelihood of an unfavorable prognosis. This suggests that homogeneous enhancement could potentially serve as a favorable prognostic indicator, reflecting a clinically meaningful trend.

Despite the valuable insights gained from our study, several limitations must be acknowledged. Firstly, the relatively small sample size and diversity of cases in our cohort limited the number of cases in each disease category, potentially affecting the generalizability of our findings. Secondly, while we employed matched pathology results and survival data from previous literature as a gold standard for prognostic classification, discrepancies between these estimates and actual patient outcomes may exist. Furthermore, many patients were lost to follow-up, which could introduce bias. Nonetheless, initial findings from follow-up data aligned with our expected prognoses based on literature-derived survival rates. Finally, since this study was retrospective and defined intratumoral hemorrhage based on imaging presentations, potential discrepancies between imaging features and anatomical findings might exist, introducing ambiguity in subgroup classification. Future investigations should explore the distinctions in patient characteristics between histologically confirmed versus imaging-defined intratumoral hemorrhage to better elucidate these differences. Moreover, larger sample sizes and prospective, multicenter studies are essential to enhance the accuracy and generalizability of prognostic assessments in pineal region tumors. Such approaches would provide more robust data for refining treatment strategies and improving patient outcomes.

Conclusions

Our study underscores the critical role of imaging features in the prognostic evaluation of pineal region tumors. Specifically, the presence of intratumoral hemorrhage emerges as a significant prognostic factor. Despite the inherent limitations of our study, such as the small sample size and retrospective nature, the identification of intratumoral hemorrhage as a poor prognostic indicator highlights the potential of advanced neuro-radiological techniques in enhancing prognostic stratification and guiding clinical management. This study emphasizes the necessity for further specialized research in neuroradiology to refine our understanding of imaging biomarkers and their predictive value in pineal region tumors.

Acknowledgments

Funding: This study was supported by the National Natural Project “To establish an automatic multimodal MRI scoring system for patients with MOYAMOYA disease through AI deep learning” (No. 82071280).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-24-281/rc

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-281/coif). The 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 (as revised in 2013). The study was approved by institutional ethics committee board of the Seventh Medical Center of Chinese PLA General Hospital (No. S2024-041-01) and individual consent for this retrospective analysis was 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. Smirniotopoulos JG, Rushing EJ, Mena H. Pineal region masses: differential diagnosis. Radiographics 1992;12:577-96. [Crossref] [PubMed]
2. Favero G, Bonomini F, Rezzani R. Pineal Gland Tumors: A Review. Cancers (Basel) 2021.
3. Jennings MT, Gelman R, Hochberg F. Intracranial germ-cell tumors: natural history and pathogenesis. J Neurosurg 1985;63:155-67. [Crossref] [PubMed]
4. Lombardi G, Poliani PL, Manara R, Berhouma M, Minniti G, Tabouret E, Razis E, Cerretti G, Zagonel V, Weller M, Idbaih A. Diagnosis and Treatment of Pineal Region Tumors in Adults: A EURACAN Overview. Cancers (Basel) 2022;14:3646. [Crossref] [PubMed]
5. Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, Hawkins C, Ng HK, Pfister SM, Reifenberger G, Soffietti R, von Deimling A, Ellison DW. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro Oncol 2021;23:1231-51. [Crossref] [PubMed]
6. Dumrongpisutikul N, Intrapiromkul J, Yousem DM. Distinguishing between germinomas and pineal cell tumors on MR imaging. AJNR Am J Neuroradiol 2012;33:550-5. [Crossref] [PubMed]
7. Fan Y, Huo X, Li X, Wang L, Wu Z. Non-invasive preoperative imaging differential diagnosis of pineal region tumor: A novel developed and validated multiparametric MRI-based clinicoradiomic model. Radiother Oncol 2022;167:277-84. [Crossref] [PubMed]
8. Li B, Feng J, Chen L, Li Y, Wang J, Lv W, Li Y, Liu Y, Liu S, Luo S, Qiu X. Relapse pattern and quality of life in patients with localized basal ganglia germinoma receiving focal radiotherapy, whole-brain radiotherapy, or craniospinal irradiation. Radiother Oncol 2021;158:90-6. [Crossref] [PubMed]
9. Broniscer A, Laningham FH, Kocak M, Krasin MJ, Fouladi M, Merchant TE, Kun LE, Boyett JM, Gajjar A. Intratumoral hemorrhage among children with newly diagnosed, diffuse brainstem glioma. Cancer 2006;106:1364-71. [Crossref] [PubMed]
10. Yamakawa K, Shitara N, Genka S, Manaka S, Takakura K. Clinical course and surgical prognosis of 33 cases of intracranial epidermoid tumors. Neurosurgery 1989;24:568-73. [Crossref] [PubMed]
11. Ostrom QT, Gittleman H, Fulop J, Liu M, Blanda R, Kromer C, Wolinsky Y, Kruchko C, Barnholtz-Sloan JS. CBTRUS Statistical Report: Primary Brain and Central Nervous System Tumors Diagnosed in the United States in 2008-2012. Neuro Oncol 2015;17:iv1-iv62. [Crossref] [PubMed]
12. Berger TR, Wen PY, Lang-Orsini M, Chukwueke UN. World Health Organization 2021 Classification of Central Nervous System Tumors and Implications for Therapy for Adult-Type Gliomas: A Review. JAMA Oncol 2022;8:1493-501. [Crossref] [PubMed]
13. Schild SE, Scheithauer BW, Haddock MG, Wong WW, Lyons MK, Marks LB, Norman MG, Burger PC. Histologically confirmed pineal tumors and other germ cell tumors of the brain. Cancer 1996;78:2564-71. [Crossref] [PubMed]
14. Wiemels J, Wrensch M, Claus EB. Epidemiology and etiology of meningioma. J Neurooncol 2010;99:307-14. [Crossref] [PubMed]
15. Hollon T, Nguyen V, Smith BW, Lewis S, Junck L, Orringer DA. Supratentorial hemispheric ependymomas: an analysis of 109 adults for survival and prognostic factors. J Neurosurg 2016;125:410-8. [Crossref] [PubMed]
16. Major K, Daggubati LC, Mau C, Zacharia B, Glantz M, Pu C. Sellar Atypical Teratoid/Rhabdoid Tumors (AT/RT): A Systematic Review and Case Illustration. Cureus 2022;14:e26838. [Crossref] [PubMed]
17. Śledzińska P, Bebyn MG, Furtak J, Kowalewski J, Lewandowska MA. Prognostic and Predictive Biomarkers in Gliomas. Int J Mol Sci 2021;22:10373. [Crossref] [PubMed]
18. Eckel-Passow JE, Lachance DH, Molinaro AM, Walsh KM, Decker PA, Sicotte H, et al. Glioma Groups Based on 1p/19q, IDH, and TERT Promoter Mutations in Tumors. N Engl J Med 2015;372:2499-508. [Crossref] [PubMed]
19. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159-74.
20. Kakigi T, Okada T, Kanagaki M, Yamamoto A, Fushimi Y, Sakamoto R, Arakawa Y, Mikami Y, Shimono T, Takahashi JC, Togashi K. Quantitative imaging values of CT, MR, and FDG-PET to differentiate pineal parenchymal tumors and germinomas: are they useful? Neuroradiology 2014;56:297-303. [Crossref] [PubMed]
21. Dixon L, Jandu GK, Sidpra J, Mankad K. Diagnostic accuracy of qualitative MRI in 550 paediatric brain tumours: evaluating current practice in the computational era. Quant Imaging Med Surg 2022;12:131-43. [Crossref] [PubMed]
22. Chen YL, Tai LH, Lieu AS. Recurrent pineal parenchymal tumor of intermediate differentiation with intratumoral hemorrhage: A case report and review of the literature. Rare Tumors 2023;15:20363613231177537. [Crossref] [PubMed]
23. Lou X, Ma L, Wang FL, Tang ZP, Huang H, Cai YQ, Wong EH. Susceptibility-weighted imaging in the diagnosis of early basal ganglia germinoma. AJNR Am J Neuroradiol 2009;30:1694-9. [Crossref] [PubMed]