Stereotactic Radiosurgery Outcomes in Medically and Surgically Failed or Nonsurgical Candidates with Medically Failed Prolactinomas: A Systematic Review and Meta-Analysis

Abstract

Objective  Prolactinomas are treated with dopamine agonists (DAs) as first-line therapy and transsphenoidal surgery as an alternative approach for medically failed tumors. We sought to summarize the efficacy of stereotactic radiosurgery (SRS) in the medically and surgically failed prolactinomas as well as in nonsurgical candidates with medically failed prolactinomas by systematic review and meta-analysis.

Method  A literature search was conducted according to the Preferred Reporting Items for Systematic Review and Meta-Analyses guideline.

Results  A total of 11 articles (total N  = 709) met inclusion criteria. Thirty-three percent of patients were able to achieve endocrine remission at a mean follow-up of 54.2 ± 42.2 months with no association between stopping DA and endocrine remission. Sixty-two percent of patients were able to achieve endocrine control with DA therapy and 34% of patients were able to decrease the dose of DA dose when compared with pre-SRS DA dose at the end of the follow-up period. However, 54% of patients required DA at the end of the follow-up to control hyperprolactinemia. Ninety percent of patients were able to achieve radiologic control at the end of the follow-up in comparison to pre-SRS imagings. Furthermore, 26% of patients newly developed hypopituitarism (one or more pituitary hormones) post-SRS throughout the follow-up period.

Conclusion  This systematic review and meta-analysis demonstrates SRS as an effective adjunct therapy in medically failed nonsurgical candidates or surgically and medically recalcitrant prolactinomas with a 33% chance of achieving endocrine remission, 62% of patients achieved hormonal control with DA and GKRS (gamma knife radio-surgery), with a 34% chance of decreasing DA dose and 90% chance of achieving radiologic control.

Introduction

Prolactinoma or prolactin (PRL) secreting pituitary adenoma or lactotroph adenoma (LA) are the most common functioning pituitary adenomas accounting for almost 40% of the pituitary adenomas. ^{1 2} Unlike other pituitary tumors, these PRL secreting adenomas are typically treated very effectively with medical therapy, with dopamine agonists (DAs) considered as first-line therapy for LAs. ^{3 4} However, 10 to 20% of the tumors do not respond adequately to DA therapy considered as DA-resistant prolactinoma, or patients develop side effects such as nausea, headache, orthostatic hypotension, psychiatric disturbances, and vertigo defined as DA intolerant. ^{4 5 6 7 8 9 10} DA resistance is defined as failure to achieve normalization of serum PRL after receiving a daily dose of bromocriptine (BRC) 15 mg or cabergoline (CAB) 1.5 to 3.0 mg weekly or equivalent for other DAs, although some advocate titration to higher but tolerable doses. ^{4 9 11 12 13 14 15} In these scenarios where medical therapy fails, surgery is considered as the viable alternative approach with transsphenoidal surgery as the first option and rarely via craniotomy.

However, surgery provides long-term remission only in 20% macroadenomas and 50% microadenomas. ^{16 17} That is because complete resection is not possible in case of tumor invading the surrounding structures, especially cavernous sinus and/or dura as trying to do so will increase the risk of venous plexus bleeding, internal carotid artery, as well as cranial nerve (3, 4, 5, 6) injury. ^{18 19} Additionally, patients who require long-term DAs to control PRL-related symptoms increase the chances of DA-related side effects. These medically and surgically failed or medically failed and surgically unamenable tumors can be approached via radiation therapy (RT). Our previous paper has reported that 8% of patients require radiotherapy after the initial surgery. ²⁰ Stereotactic radiosurgery (SRS) is preferred over conventional radiotherapy (CRT) because of its convenience, rapid correction of PRL hypersecretion, lower risk of radiation-induced neoplasms, optic neuropathy, hypopituitarism, cerebrovascular accidents, neurological disorders, and carotid stenosis. ^{21 22 23 24 25} Radiosurgical treatment may deliver a higher dose to adenoma with accuracy and may not influence or damage nearby vital structures. ²⁶

These medically as well as surgically failed prolactinomas are interesting topics to debate and therefore few individual studies are available discussing SRS outcomes in these unamenable tumors. ^{27 28 29 30 31 32 33 34 35 36 37} Our main purpose was to perform the meta-analysis of SRS outcomes of medically and surgically failed as well as medically failed prolactinomas in nonsurgical candidates to establish its efficacy in such groups of patients.

Method

Literature Search and Selection Criteria

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines. ³⁸ A detailed electronic search using Ovid Medline/PubMed, Ovid Embase, Ovid Scopus, and Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, was performed using keywords: “prolactinoma” or “pituitary adenoma” and “medically resistant” or “surgery” or “systematic review” or “meta-analysis” from their dates of inception to February 22, 2021. The search strategy was designed by a master's level librarian with inputs from the study's principal investigator ( Supplementary Material , available in the online version). Title and abstract screening followed by full-text evaluation were performed independently by two study coauthors (K.Y. and J.V.G.). The following were used as inclusion criteria: (1) cohort studies or clinical trials including patients with either DA resistant or DA intolerant grouped prolactinomas which may or may not have underwent surgery and required SRS during the time span of 2006 to 2021 and (2) studies reporting hormonal and/or radiological outcomes following SRS at last follow-up. Studies using nonhuman species, case reports, reviews, surveys, and letters to the editor were excluded.

Definitions and Outcomes of Interests

DA resistance was defined as a failure to achieve normal PRL levels after receiving a daily dose of BRC 15 mg or CAB 1.5 to 3.0 mg weekly or equivalent of other DAs. DA intolerance was defined as an occurrence of side effects that made it necessary to discontinue DA treatment. For the purposes of this article, patients with DA resistance and DA intolerance were broadly categorized as “DA failure.” Surgically failed prolactinomas were defined as prolactinomas failed to achieve normal PRL level after surgery, or recurrence of PRL elevation once being recorded normal after surgery or tumor growth recorded on imaging.

Endocrine/hormonal outcomes were subdivided into four groups: (1) Endocrine remission was defined as achieving normal PRL level without any further intervention including no DA after SRS at the last follow-up point. (2) Endocrine control was defined as achieving endocrine remission and/or normalization of PRL on DA therapy after SRS at the last point of follow-up. Although there were few exceptions in these definitions as Kara et al ²⁷ and Hung et al ²⁸ have also defined hormonal control with achieving ≤ 30 ng/mL PRL level on DA therapy after SRS at last follow-up in addition to the above definition. Tinnel et al ³⁶ have not mentioned that patients who achieved normal PRL were either on DA or off DA. So, to avoid major bias we included them in the endocrine control group rather than the remission group. Wilson et al ³² have defined achieving a hormone level < 500 µg/L at the last follow-up with or without DAs as the hormonal control group. (3) Patients who need DA treatment to control hyperprolactinemia-related symptoms or high abnormal PRL levels at the time of the last follow-up were defined as patients on DA therapy. (4) Patients who had abnormal PRL but were able to remain off the DA or were able to tolerate decreased dose of DA therapy than the pre-SRS dose at the last follow-up were defined as patients decreasing DA therapy. This post-SRS DA therapy was determined on the basis of individual patient's tolerance of serum PRL level post-SRS.

Radiological outcomes were reported as radiological control group: Radiological control was defined as either decreased size of tumor from pre-SRS size (Regression) or no change in the size of tumor from pre-SRS size (Stable) at the last follow-up point. However, there was heterogeneity in definitions of radiologic outcomes throughout different studies. Kara et al, ²⁷ Hung et al, ²⁸ Jezková et al, ³⁴ and Pouratian et al ³⁷ have defined criteria as 20% rule. As per > 20% reduction in size was defined as radiologic regression and within 20% of size was defined as radiologic stable. Similarly, Wilson et al ³² and Cohen-Inbar et al ³⁰ defined a rule of 25 and 10%, respectively. Others did not define any specific demarcation for the above.

Adverse outcomes were subdivided into three groups: (1) New hypopituitarism was defined as a newly reported abnormally low level of one or more pituitary hormones after SRS at last follow-up that may or may not require hormonal substitution. (2) Visual complications are defined as any newly worsening in vision after SRS at the last follow-up point. (3) Rest of the complications were covered under the name of other complications if reported. Requirement of another procedure was subdivided into two groups: (1) Requirement of surgery after SRS at any point of time during follow-up was defined as another surgery required. (2) Requirement of radiotherapy (SRS/CRT) at any point of time after SRS during the follow-up period was defined as another radiotherapy required. Peri-SRS DA stoppage is defined as DA off before the SRS for a specific time period. This time period varies as per individual study with a minimum of 2 weeks and a maximum of 8 weeks before SRS.

Data Extraction and Critical Appraisal

Following data were extracted from included studies: author, year of publication, study design (retrospective or prospective, observational, or trial), country, number of institutions and surgeons, total number of patients, median age of cohorts, sex distribution, median number of follow-up periods, total number patients took DA before SRS, total number patients underwent surgery before SRS, number of surgeries underwent, total number of patients on DA at the time of SRS, total number of tumors invading cavernous sinus, median PRL level before SRS, median size of tumor before SRS, total number of patients underwent SRS, type of SRS, median maximum dose of radiation, median marginal dose of radiation, median maximum point dose at optic apparatus, endocrine/hormonal outcomes, radiological outcomes, adverse outcomes, and requirement of another procedure.

Risk of bias was evaluated using the Newcastle–Ottawa Scale (NOS) for observational studies. ³⁹ The Grade of Recommendation, Assessment, Development and Evaluation (GRADE) approach was used to assess limitations in study design, consistency, evidence directness, the precision of results, and publication bias. ⁴⁰ The GRADEpro guideline development tool was used to create a summary of findings table ( http://gradepro.org/ ).

Statistical Analysis

All statistical analyses were performed using STATA 17. We performed a descriptive analysis of demographics and study characteristics. Categorical data were presented with percentages and proportions. Numerical data with a normal distribution were presented as means and standard deviations. Effect size for binary outcomes was calculated to pool effect estimates. The I ² statistic was used to determine the percentage of total variation across studies secondary to heterogeneity rather than chance, with values greater than 50% representing substantial heterogeneity. ⁴¹ Standard deviation of change scores were calculated using the correlation coefficient methodology per the suggestion of Cochrane Review Handbook. In cases of zero events for binary outcomes in the treatment arm, a continuity correction of 0.1 was applied. Fixed effects model was used when I ² < 50%. All p -values were two-sided. Level of statistical significance was established at < 0.05. Publication bias was assessed through evaluation of asymmetry on funnel plots. Sensitivity analyses were performed by omission of each study and reevaluation of the overall trend direction.

Results

Literature Search and Study Characteristics

A total of 611 results were obtained from the initial electronic search. Following the initial title, abstract screening, and full-text evaluation, 11 observational studies with a total of 709 patients were identified who underwent SRS for medically and surgically failed or nonsurgical candidates with medically failed prolactinomas. ^{27 28 29 30 31 32 33 34 35 36 37} Out of these 709 patients, 700 patients were evaluated for radiological outcomes and 695 patients were evaluated for endocrine outcomes, because Pouratian et al ³⁷ have evaluated 28 patients for radiological outcomes while 23 patients for endocrine outcomes from a total of 37 patients. Fig. 1 shows the PRISMA article search strategy. Median follow-up ranged between 6 to 140 months with a mean follow-up of 64 ± 35.3 months. The mean age of patients at the time of SRS was 39.7  ±  2.6 years. The median frequency distribution for males and females was 13.5 (3, 119) and 19 (10, 170), respectively. Based on 7 studies ( N  = 342), 96.5% of patients had prior DA exposure, while based on 8 studies ( N  = 507), 60.6% of patients had at least one prior surgery and only 9 patients in total had prior RT (SRS/CRT/fractionated stereotactic radiotherapy [FSRT]). Note that 53.2% of patients were on DA at the time of SRS ( N  = 496). Based on 7 studies ( N  = 457), 45.7% had cavernous sinus invasion. The mean size of the tumor at the time of SRS ( N  = 483) was 2.03 ± 0.8 cm ³ . The mean margin dose was 24.1 ± 5.8 Gy reported by all 11 studies ( N  = 700), while the mean maximum dose was 54.5 ± 11.3 Gy reported by only 6 studies ( N  = 440). As reported by 10 studies ( N  = 524), the mean time to reach endocrine remission was 54.2 ± 42.2 months. Study characteristics are presented in Tables 1 and 2 .

Fig 1.

Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) search strategy.

Table 1. Study characteristics.

Author	Study type	Country	Multi-institution	Total N	Male	Female	Median age (y)	Median follow-up (mo)	Prior DA	Prior surgery	Prior radiotherapy
Kara et al, 2021	Retrospective observational	Turkey	No	52	22	30	38	94.5	NA	52	NA
Hung et al, 2019	Retrospective observational	Multi (Primary USA)	Yes	289	119	170	36	60	NA	173	NA
Ježková et al, 2019	Retrospective observational	Czech Republic	No	28	15	13	37.5	140	28	9	NA
Wilson et al, 2015	Retrospective observational	Australia	No	13	3	10	38	6.1	NA	NA	NA
Cohen-Inbar et al, 2015	Retrospective observational	USA	No	38	16	22	43	42.3	38	32	4
Tanaka et al, 2010	Retrospective observational	USA	No	22	12	10	42	60	22	9	1
Wan et al, 2009	Retrospective observational	China	Yes	176	NA	NA	NA	67.3	176	NA	NA
Jezková et al, 2009	Retrospective observational	Czech Republic	No	35	8	27	40.3	66	35	10	NA
Castinetti et al, 2009	Retrospective observational	France	No	15	NA	NA	NA	96	15	7	NA
Tinnel et al, 2008	Retrospective observational	USA	No	4	NA	NA	NA	19.5	NA	NA	NA
Pouratian et al, 2006	Retrospective observational	USA	No	23 (E); 28 (R)	11 (E); 12 (R)	12 (E); 16 (R)	42.9 (E); 43.1 (R)	52 (E); 52 (R)	13 (E); 16 (R)	19 (E); 24 (R)	4 (E); 4 (R)

Abbreviations: DA, dopamine agonist; E, endocrine outcomes; NA, not available; R, radiologic outcomes.

Table 2. Study characteristics.

Author	On DA at SRS	Cavernous sinus invasion	Tumor size before SRS (cm 3 )	Radiosurgery type	Median maximum dose (Gy)	Median marginal dose (Gy)	Median max point dose (Gy) at optic apparatus	Median time for endocrinal remission (mo)
Kara et al, 2021	18	21	2.6	GKRS	NA	17	5	72.3
Hung et al, 2019	186	125	2.0	GKRS	44	22	7.1	34
Ježková et al, 2019	6	16	1.15	GKRS	70	35	5	152
Wilson et al, 2015	NA	6	1.7	XKRS	NA	20	< 8	24
Cohen-Inbar et al, 2015	21	24	2.42	GKRS	50	25	4.1	12
Tanaka et al, 2010	3	13	NA	GKRS	50	25	< 12	34
Wan et al, 2009	NA	NA	NA	GKRS	NA	22.4	< 10	NA
Jezková et al, 2009	13	NA	0.98	GKRS	70	35	< 8	96
Castinetti et al, 2009	NA	4	NA	GKRS	NA	26	NA	75
Tinnel et al, 2008	1	NA	NA	GKRS	NA	19	NA	18
Pouratian et al, 2006	13 (E); 16 (R)	NA	3	GKRS	42.2 (E); 43.1 (R)	18.6 (E); 18.9 (R)	3.6	24.5 (E)

Abbreviations: DA, dopamine agonist; E, endocrine outcomes; GKRS, gamma knife radiosurgery; NA, not available; R, radiologic outcomes; SRS, stereotactic radiosurgery; XKRS, X-knife radiosurgery.

Endocrinological/Hormonal Outcomes

Endocrine Remission

From an analysis of 9 studies ( N  = 678), 33% of patients were able to achieve endocrine remission at the end of the follow-up period after SRS (95% confidence interval [CI]: 25–41%, p  < 0.01, I ²  = 76.10%) ( Fig. 2A ).

Fig 2.

Forest plot summarizing the effect size of ( A ) endocrine remission achieved after stereotactic radiosurgery (SRS) at the last follow-up point, ( B ) endocrine control achieved after SRS at the last follow-up point, ( C ) patients on dopamine agonist (DA) therapy at the last follow-up point, and ( D ) patients decreasing DA therapy at the last follow-up period.

Endocrine Control

From an analysis of 9 studies ( N  = 496), 62% of patients were able to achieve endocrine control after SRS at the end of the follow-up period (95% CI: 47–77%, p  < 0.01, I ²  = 89.32%) ( Fig. 2B ).

Patients on DA Therapy

From an analysis of 6 studies ( N  = 165), 54% of patients required DA therapy after SRS at the end of the follow-up period to control hyperprolactinemia-related symptoms or high abnormal PRL levels (95% CI: 38–69%, p  < 0.01, I ²  = 78.46%) ( Fig. 2C ).

Patients Decreasing DA Therapy

From an analysis of 5 studies ( N  = 113), 34% of patients were able to either remain off the DA or were able to tolerate decreased dose of DA after SRS at the end of the follow-up period (95% CI: 19–50%, p  = 0.01, I ²  = 70.66%) ( Fig. 2D ).

Radiological Outcomes

Radiological Control

From an analysis of 6 studies ( N  = 342), 90% of patients were able to achieve radiologic control of tumor after SRS at the end of the follow-up period (95% CI: 87–93%, p  < 0.01, I ² = 4.42%) ( Fig. 3 ).

Fig 3.

Forest plot summarizing the effect size of radiologic control achieved after stereotactic radiosurgery (SRS) at the last follow-up point.

Furthermore, 2 studies ( N  = 32) have reported that all the patients (100%) were able to achieve radiologic control after SRS at the end of the follow-up period. ^{29 36}

Adverse Outcomes

New Hypopituitarism

From an analysis of 9 studies ( N  = 681), 26% of patients developed abnormally low levels of at least one or more pituitary hormones at the end of the follow-up period after SRS (95% CI: 14–38%, p  < 0.01, I ²  = 94.49%) ( Fig. 4A ).

Fig 4.

Forest plot summarizing the effect size of ( A ) new hypopituitarism developed after stereotactic radiosurgery (SRS) throughout the follow-up period, and ( B ) patients off dopamine agonist (DA) before SRS.

Visual Complications

From an analysis of 6 studies ( N  = 457), there were no significant visual complications after SRS throughout the follow-up period ( p  = 0.91, I ²  < 0.01%).

Other Complications

From an analysis of 3 studies ( N  = 211), we did not find any significant results throughout the follow-up period ( p  = 0.77, I ² < 0.01%).

Peri-SRS DA Stoppage

From an analysis of 4 studies ( N  = 133), stoppage of DA has no effect on achieving endocrine remission at the end of the follow-up period ( p  = 0.85, I ²  = 0%) ( Fig. 4B ).

Requirement of Another Procedure

Requirement of Another Surgery

From an analysis of 4 studies ( N  = 288), we found no effect on patients who underwent another surgery after SRS throughout the follow-up period ( p  = 0.30, I ²  = 17.82%).

Requirement of Another Radiotherapy (SRS/CRT/FSRT)

From an analysis of 3 studies ( N  = 249), we found no effect on significance for patients who required another radiotherapy after SRS throughout the follow-up period ( p  = 0.33, I ²  = 10.96%).

Critical Appraisal

Risk of Bias

Study quality assessment using the NOS revealed high quality/low risk of bias for seven observational studies and moderate quality/moderate risk of bias for four observational studies ( Supplementary Table S1 , available in the online version). Funnel plots were generated for publication bias and no apparent asymmetry was found on visual inspection.

Strength of Evidence

Strength of evidence assessed, using a GRADE approach, was moderate for endocrine remission, radiologic regression, and new hypopituitarism, low for endocrine control and radiological stable, and very low for patients on DA therapy, patients decreasing DA therapy, radiologic control, visual complications, other complications, requirement of another surgery, and requirement of another radiotherapy. The GRADE summary of findings is summarized in Supplementary Table S2 (available in the online version).

Discussion

This is the meta-analysis on SRS outcomes on medically resistant and/or intolerant prolactinomas with mainly endocrine/hormonal, radiologic, and adverse outcomes. We report SRS as last resort—after considering DA and surgery—successfully achieved 33% endocrine remission, 62% endocrine control, and 34% were able to decrease DA dose at the end of follow-up period, therefore only 4% of patients did not get any level of hormonal success after SRS. Additionally, 90% achieved radiologic control after SRS in medically failed and surgically failed or surgically nonresectable prolactinomas. Hormone remission rate and radiologic control rate are being supported by a very recent article. ⁴²

DA is without a doubt the initial choice of treatment for prolactinomas and surgery—mainly transsphenoidal resection—is a second line therapy in cases of DA resistant or DA intolerant patients. After failure of first liners, patients look for alternate approaches like continuation of DA for lifetime to suppress hyperprolactinemia-related symptoms, or another surgery to resect recurrence or residual tumor. However, when a tumor is invading surrounding structures like cavernous sinus or dura, complete surgical resection is not possible and therefore, patients end up undergoing multiple surgeries or may require lifetime DA. Either option is less feasible as long-term medicine may induce DA-related side effects including some psychological disorders ^{4 5 9 43} as well as cardiac valvular thickening and regurgitation, ^{44 45} and surgery in such cases, there are higher chances of cranial nerve injuries, internal carotid artery injury, or venous plexus bleeding. ^{4 5 9 18 19} In such scenarios, radiotherapy is an alternate or adjuvant option to achieve hormonal control and/or tumor control. CRT was the initial choice but because of higher morbidity and mortality related to CRT as well as nowadays because of availability of other methods like fractionated radiotherapy or SRS that can deliver higher dose of radiation to the tumor and lower dose to the surrounding normal tissue, CRT is not an option of choice for the same. SRS has emerged as the choice of therapy since the past two decades because of its convenience, rapid correction of hormonal level, and lesser chances of radiation-related side effects. ^{21 22 23 24} However, there are certain limitations to the SRS like tumor size and proximity of the tumor with the optic apparatus, distance should be at least 2 mm between tumor and optic apparatus. ⁴⁶ In such cases, fractionated external beam radiotherapy (FEBRT) is preferred as maximum radiation dose in FEBRT is usually in tolerable range for the brain. ⁴⁶

Cornerstone outcomes in this present systematic review and meta-analysis are endocrine outcomes. Thirty-three percent of patients successfully achieved endocrine remission without further need of any treatment and 62% of patients achieved hormonal control with DA as adjuvant therapy. Moreover, 34% of patients were able to decrease the dose of the DA. Additionally, Wilson et al, ³² Tanaka et al, ³¹ Jezková et al, ³⁴ and Castinetti et al ³⁵ have shown that patients were also able to completely cease the medication post-SRS even with high PRL level. Jezková et al ³⁴ and Ježková et al ²⁹ also reported that female patients who were not able to become pregnant because of DA have achieved pregnancy after SRS either as a result of complete cessation of DA or decreasing dose of DA. However, to control hyperprolactinemia 54% of patients were on DA even after SRS. Furthermore, talking about DA use before SRS, a topic to debate, some studies have reported stopping using DA before SRS is associated with achieving endocrine remission, ^{35 37 47 48} while others have reported that there is no association between stop DA usage before SRS. ^{29 30 34 49} Our present systematic review and meta-analysis support the later reporting there is no association between DA off before SRS. Many other studies have reported pre-SRS PRL level and pre-SRS tumor size is also good predictor of post-SRS hormonal remission. ^{28 31 35 37} Even for medical and surgical therapy, baseline PRL value and tumor sizes are prognostic factors and predictors of tumor invasiveness and hormonal outcomes. ^{50 51 52 53 54}

Because prolactinomas are PRL secreting tumors and patients mainly seek attention because of signs and symptoms related to hyperprolactinemia, endocrine outcomes are primarily considered as markers of treatment failure or success nevertheless tumor regression or growth control. In our study we have reported radiologic outcomes to check SRS effect on tumor volume, 90% of patients reported tumor control. Also studies by Tinnel et al ³⁶ and Ježková et al ²⁹ have reported 100% tumor control rate. Proving SRS as a comprehensive therapy successfully controls the volume of refractory prolactinomas or maybe even complete disappearance of tumors. ^{29 34} Pouratian et al ³⁷ reported no patient characteristics, no tumor characteristics and no SRS characteristics are associated with tumor control, while Kara et al ²⁷ have reported hormonal control and tumor control are associated in positive correlation and K _i -67 proliferative index, as well as cavernous sinus invasion, are negatively correlated with the tumor control.

Adverse outcomes are always an interesting topic for radiotherapy. Most important side effect after SRS on the pituitary is the development of hypopituitarism with at least one pituitary hormone, previously reported up to 42% in literatures. ^{35 49} In our systematic review and meta-analysis we have reported 26% of patients developed new hypopituitarism after SRS. This is lower than FEBRT, with reportedly 50% chances of developing new hypopituitarism after FEBRT and higher chances of getting temporal lobe necrosis. ^{55 56 57} These are due to radiation-induced damage to the normal gland, hence we suggest delivering radiation within the maximum safety dose to the pituitary (< 15 Gy) and distal infundibulum (< 17 Gy). ^{58 59 60} Apart from the newly developed hypopituitarism after SRS, we have not found any complications that are significantly associated with the SRS. Although few studies have individually reported visual changes, cerebrospinal fluid leak, and radiation necrosis of surrounding structure. We suggest the same as delivering radiation within the maximum safety limits to the individual structures like < 8 Gy to the optic pathway and < 14Gy to the brainstem. However, few studies have reported some portion of the optic pathway can tolerate as high as 12 Gy with very less chances of injury. ^{61 62 63 64} Another approach for tumor encroaching under 1 mm of the optic pathway, we can initially approach for surgical resection and adjuvant SRS within 8 weeks of resection or fractionated delivery of SRS.

Limitations

There were some important limitations to the present systematic review and meta-analysis. Due to the observational nature of all included studies, we observed a very low, low, and moderate strength of evidence for outcomes on GRADE evaluation and the quality of the study was also moderate on NOS for four studies. Notably, outcomes in various studies had heterogeneous definitions. Additionally, all outcomes of interest were not reported consistently in all included studies. Duration of follow-up varied considerably (6 to 140 months) and likely had an impact on outcomes. Moreover, there was very high heterogeneity ( I ² value) for outcomes reported. Hospital and procedure costs, as well as resource utilization, were not evaluated in the included studies. It was also not possible to exclude the effect of heterogeneity in the type of DA, surgical, as well as SRS technique on outcomes. In addition, given that this was a meta-analysis, the effect of confounders such as age, sex, other comorbidities, and prolactinoma subdivision could not be adjusted via multivariable analysis.

Conclusion

Our present systematic review and meta-analysis on SRS outcomes of medically and surgically failed prolactinomas, as well as medically failed prolactinomas in nonsurgical candidates, has demonstrated 33% of patients achieved endocrine remission without the need for any further therapy in the mean time of 54.2 ± 42.2 months, 62% of patients achieved endocrine control with post-SRS DA therapy at the end of the follow-up period. Furthermore, there is no association between DA stoppage before SRS and endocrine remission. Thirty-four percent of patients were able to decrease dose DA after SRS at the end of the follow-up period, 90% of patients achieve radiologic control at the end of the follow-up period. In addition, 26% of patients developed new hypopituitarism as an adverse effect of SRS therapy throughout the follow-up period as well as 54% of patients required DA therapy at the end of the follow-up period to control hyperprolactinemia and/or symptoms related to hyperprolactinemia. Hence, SRS is not the initial choice for prolactinomas as potential side effects related to it but can be very promising in selective candidates who have failed both medicine and surgery or have failed medicine but surgery is not possible because of tumor location and invasion.