Gamma Knife radiosurgery for the management of nonfunctioning pituitary adenomas: a multicenter study
Jason P. Sheehan, Robert M. Starke, David Mathieu, Byron Young, Penny K. Sneed, Veronica L. Chiang, John Y. K. Lee, Hideyuki Kano, Kyung-Jae Park, Ajay Niranjan, Douglas Kondziolka, Gene H. Barnett, Stephen Rush, John G. Golfinos, L. Dade Lunsford J Neurosurg. 2013 Aug;119(2):446-56. doi: 10.3171/2013.3.JNS12766
Context:
Pituitary adenomas account for between 10 and 20% of all intracranial tumours. Of these, 15-30% are nonfunctioning adenomas, tumours that do not secrete an active pituitary hormone. Symptoms are produced through mass effect on the optic chiasm leading to optic field or acuity loss, or on the pituitary gland leading to hypopituitarism. The usual management of symptomatic or enlarging nonfunctioning pituitary ademoas normally undergo surgical resection. Recurrence has been reported as between 24% and 80%. These patients often have evidence of invasion of the cavernous sinus or dural invasion. Stereotactic radiosurgery (SRS) is an important management option in recurrent pituitary adenomas, and less commonly as the initial management if the patient has high surgical comorbidities. This study aimed to examine the outcomes after radiosurgery in a large, multicentre patient population in the USA.
Methods:
Nine medical centres affiliated with the North American Gamma Knife Consortium submitted their retrospective clinical outcome analysis of 512 patients with nonfunctioning pituitary adenomas between 1988 and 2011. Information was extracted from the case notes and screened by an independent third party. Clinical and imaging outcomes were assessed at a median follow up period of 36 months (range 1-223 months) after stereotactic radiosurgery. Mean age was 53 years, 44% were females, 56% males, 93.6% underwent at least one prior resection prior to SRS, 41.5% had more than one precious surgical resection, 3.6% had prior fractionated radiation therapy. All patients were assessed serially by clinical, cranial nerve and neuroendocrine evaluation. Median tumour volume was 3.3cm3 (range 0.08-35.2cm cm3), a median of 8 isocentres was used for dose planning, the median prescription dose delivered to the tumour margin was 16G, the maximum dose varied from 10-70 Gy. Follow up was 6 monthly for the first 1-2 years, then every 1-2 years by the treating neurosurgeon or the referring physicians. Follow up images were compared to those obtained at the time of SRS, the volume calculated roughly by multiplying the AP, lateral and superior-inferior and dividing the number by 2. Tumour growth withing the prescribed isodose volume or adjacent to it was considerd tumour progression. Statistical tests used were chi-quared, Fisher exact, Student t-test and Mantel-Haenszel tests. Kaplan-Meier curves were plotted for survival and progression free survival and overall survival from the first SRS date. Univeriate analysis was performed on the Kaplan Meier curves using log rank statistics.
Results:
Median follow up was 36 months. At last follow up 31 (6.6%) of 469 patients with available follow up and imaging had tumour progression. Overall tumour control was 93.4%. Actuarial progression-free survivals at 3, 5, 8, 10 years were 98%, 95%, 91% and 85% respectively. There was no significant difference in the progression free survival in patients with upfront radiosurgery versus those who had a prior resection.
Tumour volume (<5cm3 v >5cm3) and suprasellar extension were significant predictors of tumour progression. Those treated with 16 Gy or greater were less likely to demonstrate tumour progression.
41 (9.3%) of 442 patients had a new or worsening cranial nerve deficit. Patients with tumour progression despite SRS were more likely to have new or worsening cranial nerve deficits. 29 (6.6%) of 442 patients had new or worsening optic nerve dysfunction. Cranial nerve deficits were more common in patients with larger tumours, in patients with a history of prior hypopituitarism or prior radiation therapy and in younger patients.
92 of 435 patients developed new or worsening pituitary dysfunction. Those with preoperative endocrine dysfunction (21.2%) were NOT more likely to develop worsening pituitary dysfunction post SRS. Increasing margin dose and history of prior radiotherapy were predictive of new or worsening pituitary dysfunction. 451 (88%) of 512 achieved a favourable outcome defined as no tumour regrowth and preservation of neurological function. Factors related to favourable outcome were age older than 50, tumour volume less than 5cm3 and no prior radiotherapy.
The authors proposed a Radiosurgical Grading System: 1 point for > 50 years old, 0 points for ≤ 50 years old; tumor volume: 1 point for tumor volume < 5 cm3, 0 points for tumor volume ≥ 5 cm3; and prior radiation: 2 points for no previous radiation, 0 points for prior radiation. Favorable outcomes for the RPS were as follows: RPS of 4 had a favorable outcome of 95%; RPS of 3 had a favorable outcome of 88%; RPS of 2 had a favorable outcome of 67%; RPS of 1 had a favorable out- come of 50%; and RPS of 0 had a favorable outcome of 20%.
Comment:
The difficulty in achieving a complete resection of nonfunctioning pituitary adenomas and the tendency for them to recur makes the study of alternatives to further surgery highly relevant. One alternative, entire brain radiation therapy (EBRT), is associated with a 20-40% risk of hypopituitarism, a 2% risk of radiation induced neoplasia at 10 years and a significantly increased risk of cerebrovascular deaths. SRS is a widely used alternative to EBRT, but has not been studied other than in single centre retrospective studies. This report is the first multicentre retrospective study to date. It demonstrates a high rate of tumour control from using SRS for nonfunctioning pituitary adenomas (93.4%) and that smaller tumours and those without suprasellar extension have greater progression free survival. Based on this the authors propose a role for cytoreductive surgery in particular to reduce the size of the suprasellar component prior to SRS.
They note poorer tumour control if less than 16 Gy is given to the tumour margin. This is relevant as often doses of 12 Gy or less are given to tumour portions adjacent to the optic nerve. The authors recommend that if it is not possible to give more than 15 Gy then repeat microsurgery should be considered.
Mean time to tumour volume reduction was 33 months, exceeding mean follow up time in most previous series, therefore the authors recommend follow up to assess long term radiological, endocrine and neurological complications of 20 years or more.
Pituitary adenomas account for between 10 and 20% of all intracranial tumours. Of these, 15-30% are nonfunctioning adenomas, tumours that do not secrete an active pituitary hormone. Symptoms are produced through mass effect on the optic chiasm leading to optic field or acuity loss, or on the pituitary gland leading to hypopituitarism. The usual management of symptomatic or enlarging nonfunctioning pituitary ademoas normally undergo surgical resection. Recurrence has been reported as between 24% and 80%. These patients often have evidence of invasion of the cavernous sinus or dural invasion. Stereotactic radiosurgery (SRS) is an important management option in recurrent pituitary adenomas, and less commonly as the initial management if the patient has high surgical comorbidities. This study aimed to examine the outcomes after radiosurgery in a large, multicentre patient population in the USA.
Methods:
Nine medical centres affiliated with the North American Gamma Knife Consortium submitted their retrospective clinical outcome analysis of 512 patients with nonfunctioning pituitary adenomas between 1988 and 2011. Information was extracted from the case notes and screened by an independent third party. Clinical and imaging outcomes were assessed at a median follow up period of 36 months (range 1-223 months) after stereotactic radiosurgery. Mean age was 53 years, 44% were females, 56% males, 93.6% underwent at least one prior resection prior to SRS, 41.5% had more than one precious surgical resection, 3.6% had prior fractionated radiation therapy. All patients were assessed serially by clinical, cranial nerve and neuroendocrine evaluation. Median tumour volume was 3.3cm3 (range 0.08-35.2cm cm3), a median of 8 isocentres was used for dose planning, the median prescription dose delivered to the tumour margin was 16G, the maximum dose varied from 10-70 Gy. Follow up was 6 monthly for the first 1-2 years, then every 1-2 years by the treating neurosurgeon or the referring physicians. Follow up images were compared to those obtained at the time of SRS, the volume calculated roughly by multiplying the AP, lateral and superior-inferior and dividing the number by 2. Tumour growth withing the prescribed isodose volume or adjacent to it was considerd tumour progression. Statistical tests used were chi-quared, Fisher exact, Student t-test and Mantel-Haenszel tests. Kaplan-Meier curves were plotted for survival and progression free survival and overall survival from the first SRS date. Univeriate analysis was performed on the Kaplan Meier curves using log rank statistics.
Results:
Median follow up was 36 months. At last follow up 31 (6.6%) of 469 patients with available follow up and imaging had tumour progression. Overall tumour control was 93.4%. Actuarial progression-free survivals at 3, 5, 8, 10 years were 98%, 95%, 91% and 85% respectively. There was no significant difference in the progression free survival in patients with upfront radiosurgery versus those who had a prior resection.
Tumour volume (<5cm3 v >5cm3) and suprasellar extension were significant predictors of tumour progression. Those treated with 16 Gy or greater were less likely to demonstrate tumour progression.
41 (9.3%) of 442 patients had a new or worsening cranial nerve deficit. Patients with tumour progression despite SRS were more likely to have new or worsening cranial nerve deficits. 29 (6.6%) of 442 patients had new or worsening optic nerve dysfunction. Cranial nerve deficits were more common in patients with larger tumours, in patients with a history of prior hypopituitarism or prior radiation therapy and in younger patients.
92 of 435 patients developed new or worsening pituitary dysfunction. Those with preoperative endocrine dysfunction (21.2%) were NOT more likely to develop worsening pituitary dysfunction post SRS. Increasing margin dose and history of prior radiotherapy were predictive of new or worsening pituitary dysfunction. 451 (88%) of 512 achieved a favourable outcome defined as no tumour regrowth and preservation of neurological function. Factors related to favourable outcome were age older than 50, tumour volume less than 5cm3 and no prior radiotherapy.
The authors proposed a Radiosurgical Grading System: 1 point for > 50 years old, 0 points for ≤ 50 years old; tumor volume: 1 point for tumor volume < 5 cm3, 0 points for tumor volume ≥ 5 cm3; and prior radiation: 2 points for no previous radiation, 0 points for prior radiation. Favorable outcomes for the RPS were as follows: RPS of 4 had a favorable outcome of 95%; RPS of 3 had a favorable outcome of 88%; RPS of 2 had a favorable outcome of 67%; RPS of 1 had a favorable out- come of 50%; and RPS of 0 had a favorable outcome of 20%.
Comment:
The difficulty in achieving a complete resection of nonfunctioning pituitary adenomas and the tendency for them to recur makes the study of alternatives to further surgery highly relevant. One alternative, entire brain radiation therapy (EBRT), is associated with a 20-40% risk of hypopituitarism, a 2% risk of radiation induced neoplasia at 10 years and a significantly increased risk of cerebrovascular deaths. SRS is a widely used alternative to EBRT, but has not been studied other than in single centre retrospective studies. This report is the first multicentre retrospective study to date. It demonstrates a high rate of tumour control from using SRS for nonfunctioning pituitary adenomas (93.4%) and that smaller tumours and those without suprasellar extension have greater progression free survival. Based on this the authors propose a role for cytoreductive surgery in particular to reduce the size of the suprasellar component prior to SRS.
They note poorer tumour control if less than 16 Gy is given to the tumour margin. This is relevant as often doses of 12 Gy or less are given to tumour portions adjacent to the optic nerve. The authors recommend that if it is not possible to give more than 15 Gy then repeat microsurgery should be considered.
Mean time to tumour volume reduction was 33 months, exceeding mean follow up time in most previous series, therefore the authors recommend follow up to assess long term radiological, endocrine and neurological complications of 20 years or more.