Stereotactic radiosurgery for tumors of the pineal region: A single-center experience

Abstract

Tumors of the pineal region (TPRs) are rare neoplasms that are surgically challenging to resect. Conventional treatment strategies are available, but gamma knife radiosurgery (GKRS) is an alternative approach. This study presents a single-center experience with GKRS performed for TPR with and without histopathological diagnoses. The cases of 25 patients with TPRs treated with GKRS were retrospectively analyzed. Thirteen of these 25 patients had histopathological confirmation, and 13 had elevated serum alpha-fetoprotein and beta-human chorionic gonadotropin levels. The 25 patients had a mean follow-up duration of 61 months. The total response rate to GKRS was 60%, and a 53.8% decrease in the alpha-fetoprotein and beta human chorionic gonadotropin levels was observed. The findings of this study indicate that GKRS is a safe procedure for TPRs, even in the event of insufficient histopathological findings. This treatment approach provides increased Karnofsky performance scores and an extended life expectancy.

1. Introduction

Tumors of the pineal region (TPRs) represent a rare group of neoplasms, accounting for only 1% to 3% of all intracranial tumors.^[1] These tumors exhibit a wide range of histological types. The most common neoplasms of the pineal region are germ cell tumors. Following these in frequency, we find an assortment of neoplasms that includes pineal parenchymal tumors (comprising pineocytomas and pineoblastomas), metastases, neuronal tumors, meningiomas, lymphomas, ependymomas, and gliomas.^[2,3]

Different treatment strategies are available for TPRs. Radiotherapy and chemotherapy are offered for malignant tumors, while surgery is a treatment option for benign tumors. Enhanced microsurgical techniques are commonly used; nevertheless, surgery remains challenging because of the deep anatomic localization of TPRs and proximity to critical structures. Furthermore, obtaining appropriate biopsy materials is highly risky.^[4,5] More biopsy material is also needed for pathologists.^[6] When there is no histopathological confirmation of TPR, radiosurgery is an alternative treatment,^[7] but decisions to treat pineal tumors with only radiotherapy may also be made.

Germ cell tumor markers can be easily obtained from cerebrospinal fluid or serum, and the disease can be diagnosed without pathological verification. In a study conducted by Kanamori et al^[8], 38 of 39 patients with TPRs were diagnosed using cerebrospinal fluid/serum markers and magnetic resonance imaging (MRI) findings. TPR markers, such as alpha-fetoprotein (AFP) and beta-human chorionic gonadotropin (β-HCG), are also important prognostic factors, but positive markers generally only indicate malignant disease.

Gamma knife radiosurgery (GKRS) is used for TPRs, but more studies are needed on this approach in the literature.^[3,9] The pineal region is complicated because of its neighborhood with critical anatomic structures, and TPRs include many different pathological groups; for these reasons, radiotherapy and radiosurgery in the pineal region are both risky procedures.^[10] The present study was planned to share a single-center experience with GKRS for cases of TPRs with and without pathological diagnoses.

2. Methods

Ethics committee approval was waived because this study was a retrospective chart analysis. We retrospectively reviewed the data of 35 patients with TPRs treated with GKRS between 2004 and 2020 at the Gazi University Gamma Knife Center. The diagnosis was based on MRI findings, and histopathological diagnoses were made for 13 of the 25 patients. The 20 male and 15 female patients had a mean age of 37.9 (9–79) years. Patient information was obtained from Gazi University’s patient archives and the electronic records of the Turkish Ministry of Health. Ten patients who did not attend the follow-up appointments and had insufficient data were excluded from the study. For the first 12 patients treated between 2004 and 2014, the Leksell Gamma Knife Model 4 was used, and for 13 patients treated between 2014 and 2020, the Leksell Gamma Knife Perfexion was used (Table 1).

Table 1.

Summary of the patients treated with GKRS.

Case No.	Age/Sex	Tumor type/ Elevated AFP and β-HCG	Volume (cm3)	Mean diameter (mm)	GKM	Marginal dose	Prior treatment	Follow-up time	Response	Compl./Death
1	19/M	Unknown/ +	8.54	26.3	Model 4	14 Gy	V/P shunt	8 yr	CR	No
2	15/M	Unknown/ +	1.58	14.9	Model 4	15 Gy	V/P shunt	13 yr	CR	No
3	18/M	Unknown/ -	1.42	16	Model 4	14 Gy	-	3 yr	PR	No
4	20/M	Teratoma/ +	38.14	38.5	Model 4	13 Gy	Bx, V/P shunt	1 year	NR	Exitus
5	23/F	Pineoblastoma/-	0.7	11.4	Model 4	13 Gy	Surgery, V/P shunt	9 yr	CR	No
6	9/F	Unknown/ -	3.26	17.9	Model 4	12 Gy	-	2 yr	NR	Exitus
7	50/F	Pineocytoma/-	2.89	17.5	Model 4	14 Gy	Bx	4 yr	PR	No
8	16/F	Germinoma/ +	4.20	19.5	Model 4	14 Gy	V/P shunt	11 yr	PR	No
9	60/F	Meningioma/-	13.75	27.7	Model 4	14 Gy	Surgery	10 yr	PR	Brain edema
10	45/M	Pineocytoma/-	2	13	Model 4	13 Gy	Bx	11 yr	PR	No
11	21/F	Germinoma/ +	8.85	26.5	Model 4	16 Gy	Bx, RT	5 yr	NR	No
12	61/F	Unknown/ +	4.15	20.4	Model 4	15 Gy	-	7 yr	PR	No
13	50/F	Meningioma/-	11.11	27.7	Perfexion	14 Gy	Surgery	9 yr	PR	No
14	39/M	Unknown/ +	14.31	33.3	Perfexion	15 Gy	-	2 yr	PR	Brain edema
15	60/F	Meningioma	9.83	27	Perfexion	12 Gy	-	6 yr	NR	No
16	69/M	Unknown/ +	5.35	22.3	Perfexion	15 Gy	-	5 yr	PR	No
17	30/F	Unknown/ +	2.68	16.6	Perfexion	13 Gy	-	5 yr	NR	No
18	31/M	Germinoma	3.95	27.6	Perfexion	13 Gy	-	3 yr	NR	Exitus
19	62/M	Papillary tumor	2.57	16	Perfexion	14 Gy	-	3 yr	NR	No
20	62/M	Unknown/ +	1.56	14.8	Perfexion	13 Gy	-	3 yr	NR	No
21	36/F	Neurocytoma/ +	12.57	31.7	Perfexion	18 Gy	-	3 yr	NR	Exitus
22	32/M	Unknown/-	1.48	14.8	Perfexion	14 Gy	Surgery	3 yr	CR	No
23	20/M	Unknown/ +	0.9	14.5	Perfexion	15 Gy	V/P shunt, surgery	2 yr	CR	No
24	9/F	Pilocytic astrocytoma	5.54	22.1	Perfexion	12 Gy	Surgery	1 yr	PR	No
25	71/M	Unknown/ +	6.97	24.2	Perfexion	14 Gy	-	1 yr	NR	No

AFP = alpha-fetoprotein, Bx = Biopsy, Compl = complications, CR = complete remission, Gy = grey, GKM = Gamma Knife model, GKRS = gamma knife radiosurgery, β-HCG = beta human chorionic gonadotropin, NR = no remission, PR = partial remission, V/P = Ventriculoperitoneal.

The median Karnofsky performance score of the patients was 60 (range:50–80). Six patients had hydrocephalus and underwent a ventriculoperitoneal shunt surgery before GKRS. Nine patients underwent other prior surgeries, and 4 underwent open biopsies for TPR. Among these 13 patients, the histopathological results were as follows: 1 teratoma, 3 germinomas, 2 pineocytomas, 1 pineoblastoma, 1 pilocytic astrocytoma, 1 neurocytoma, 1 papillary tumor, and 3 meningiomas. One patient with germinoma received fractioned radiotherapy, and another received hypofractionated radiotherapy with 6 sessions of 6 grey (Gy) at another medical center; the patient was included in this study (see Table S1, Supplemental Digital Content, http://links.lww.com/MD/J123, which illustrates all properties of patients treated with GKRS). Twelve patients had no histopathological diagnoses because of difficulty in surgical procedures or personal preference. Three patients without histopathological diagnosis underwent ventriculoperitoneal shunting surgery.

A Leksell Gamma Knife (Elekta Instruments, Atlanta, GA) with a stereotactic frame was used for all patients under local anesthesia, and sedation was administered during the procedure. General anesthesia was administered to 2 pediatric patients. Pre-gamma knife isotropic stereotactic axial, sagittal, and coronal MRI images were reconstructed using GammaPlan 10.1.1 software (Elekta Instruments). Single- or multiple-isocenter plans were used to construct a 3D conformal radiation volume to match the tumor margins. A neurosurgeon, radiation oncologist, and medical physicist designed the treatment plans. Based on the MRI, 1 to 8 isocenters were used for treatment planning. The maximum dose was 24 to 36 Gy, and the marginal dose was 12 to 18 Gy. Patients were generally discharged 24 to 72 hours after GKRS.

As a routine procedure for each patient, clinical and radiological follow-up appointments were held at 3, 6, 12, 18, and 24 months after treatment. After 2 years, patients with stable tumors were followed-up annually. Two independent radiologists evaluated the radiological results. Based on RECIST (response evaluation criteria in solid tumors) guideline and the previous studies of Kano et al and Hasegawa et al^[3], radiological outcome classification was made for response to radiosurgery.^[6,9] According to MRI findings, we classified patients radiosurgery responses as complete remission (CR), in which tumors completely regressed; partial remission (PR), with more than 50% decrease in tumor volume; and no remission (NR), in which the tumor volumes were stable. All the follow-up data were recorded.

3. Results

Twenty-five of the 35 patients (71.4%) were followed-up regularly and were included in this study. The most common symptoms were headache and vomiting at the time of diagnosis. The mean duration from onset to the GKRS was 1.2 years. Twenty-three patients were discharged 24 to 72 hours after treatment. Two patients experienced progression of mild brain edema as a complication of the GKRS treatment. Anti-edema drugs were administered, and the patients were discharged 120 hours later. The mean follow-up duration was 61 months (1–13 years).

Thirteen patients (48%) had elevated AFP and β-HCG levels. Except for 1 patient with teratoma who died, decreases in AFP and β-HCG levels resulted in 3 cases of CR and 5 cases of PR, which may indicate positive responses to GKRS among these patients. According to the pre-GKRS MRI data, the mean tumor volume was 7.31 (0.728–26.71) cm³. CR was observed in 5 cases (20%), PR was observed in 10 cases (40%), and NR was observed in 10 cases (40%) cases, respectively. Thus, the GKRS yielded a 60% response rate in this study (see Table S2, Supplemental Digital Content, http://links.lww.com/MD/J124, which illustrates the rates of response to treatment).

CR was achieved by 5 patients with mean tumor volume of 2.64 (0.7–8.54) cm³, PR was achieved by 10 patients with mean tumor volume of 6.46 (2–13.75) cm³, and NR was achieved by 10 patients with mean tumor volume of 9.12 (1.56–38.14) cm³. Among patients with CR, the median marginal dose was 14.2 Gy. Among the patients with PR, the median marginal dose was 14.4 Gy; among the NR patients, the median marginal dose was 13.8 Gy.

The tumor progression-free survival rate in the first year was 96%. At 5 years, the rate was 84%. Four patients died within 1 to 3 years. One of these patients was diagnosed with teratoma, and 1 with extraventricular neurocytoma in the NR group. The other 2 patients died due to systemic diseases. The remaining 21 patients were alive without evidence of tumor progression or extracranial spreading. The median Karnofsky performance score of all patients with TPR was 68 (range:0–90). The median Karnofsky performance score of all patients, including the 4 deaths, increased by 8 points between the GKRS and the end of the follow-up. No further complications, pseudocyst formation, or necrosis were observed.

4. Discussion

There are standard treatment modalities for TPRs, but the response rates for each histopathological type remain unclear.^[10,11,12] GKRS is a single high-dose radiotherapy that targets the lesion using stereotactic techniques and protects the surrounding tissues as much as possible; it is an effective therapeutic modality for deep cerebral tumors.^[6,7] GKRS can be performed for TPR tumors with or without a histopathological diagnosis.^[13] In Gamma Knife treatment planning, similar doses of radiation are given to different histopathological types of pineal region tumors, which causes a debate about whether the histopathological diagnosis is necessary when making the final treatment decision; unfortunately, it cannot be provided in some cases. Advanced radiological imaging techniques can inform physicians more accurately about the nature of TPR, such as whether it is a germ cell tumor or non-germ cell tumor.^[8]

In this study, the 5-year survival rate was 60%. This rate was 66.7% in the study by Li et al^[13] and 94% in the study by Kanamori et al^[8] Furthermore, in the present study, none of the patients received further treatment, such as surgery, chemotherapy, or radiotherapy. The differences in survival rates may be attributed to pineal tumor histology, the patients personal characteristics, and their responses to GKRS. Germ cell tumors of the pineal region are more radiosensitive than other TPRs and usually progress with high AFP and β-HCG levels.^[1] In our study, the GKRS response rate in patients with elevated AFP and β-HCG levels was 53.8% (Table 2). The total response rate to the GKRS was 60%. Although the GKRS procedure can be performed without a histopathological diagnosis, the role of histopathological diagnoses in the precision of treatment and the success rate cannot be denied.^[5]

Table 2.

Number of responses to treatment and rates.

Response to GKRS	Number of patients	Mean marginal dose	High AFP and β-HCG
CR	5	14.2 Gy	3 ↓
PR	10	14 Gy	4 ↓
NR	10	13.8 Gy	6 ↔
Total response rate %	60		53.8

AFP = alpha feto protein, CR = complete remission, Gy = Grey, GKRS = gamma knife radiosurgery, β-HCG = beta human chorionic gonadotropin, NR = no remission, PR = partial remission, β-HCG = beta-human chorionic gonadotropin.

Two pineocytomas and 1 pineoblastoma were pre-diagnosed among the patients in the present study. GKRS successfully treated these patients, with 1 patient achieving CR and 2 achieving PR. One patient was diagnosed with astrocytoma, and GKRS provided a PR. Currently, GKRS is effective for low-grade astrocytomas.^[14] For the 12 patients without a histopathological diagnosis, GKRS resulted in 4 cases of CR, 4 cases of PR, and 4 cases of NR, with a response rate of 66.6%.

GKRS provides a much better quality of life than invasive techniques, such as surgery or treatment methods, with systemic side effects, such as chemotherapy and conventional radiotherapy. With GKRS performed in the pineal region, the major morbidity rate decreased to 7%, and the minor morbidity rate decreased to 30%.^[3,9,11] Most patients overall medical condition improved 6 months after GKRS. No significant complications were observed in our study after GKRS, similar to the literature findings.^[3,6,7,15] Other radiosurgery treatment methods, including CyberKnife radiosurgery, are also available for TPRs. Ahn et al^[16] published a series of 6 cases of pineal parenchymal tumors, and except for 1 patient, all achieved complete or partial remission. Frameless radiosurgery is another beneficial treatment method for local tumor control.^[17]

This study had several limitations, including the small sample size, the fact that it was a single-center study, and the need for diagnosis in some patients. Future studies should also compare the effectiveness of GKRS with other treatment modalities. Additionally, larger multicenter studies will confirm the findings of this study and determine the generalizability of the results to other patient populations.

5. Conclusion

GKRS is a safe procedure for the treatment of TPRs and a good alternative for germ cell tumors. GKRS can be applied to patients with TPR, with or without histopathological confirmation. Decision-making is possible only with a radiological diagnosis, but histopathological confirmation and elevated serum markers will make the surgeon more confident. With a low complication rate, this approach offers a successful treatment of deep central nervous system lesions.

Author contributions

Conceptualization: Mete Zeynal.

Formal analysis: Gökhan Kurt.

Methodology: Burak Karaaslan.

Project administration: Alp Börcek.

Resources: Özlem Dağli.

Supervision: Hakan Hadi Kadioğlu, Ömer Hakan Emmez.

Writing – original draft: Mete Zeynal.