Radiotherapy (RT) is often employed in patients with acromegaly refractory to medical and/or surgical interventions in order to prevent tumour regrowth and normalize elevated GH and IGF-I levels. It achieves tumour control and hormone normalization up to 90% and 70% of patients at 10–15 years. Despite the excellent tumour control, conventional RT is associated with a potential risk of developing late toxicity, especially hypopituitarism, and its role in the management of patients with GH-secreting pituitary adenomas remains a matter of debate. Stereotactic techniques have been developed with the aim to deliver more localized irradiation and minimize the long-term consequences of treatment, while improving its efficacy. Stereotactic irradiation can be given in a single dose as stereotactic radiosurgery (SRS) or in multiple doses as fractionated stereotactic radiotherapy (FSRT). We have reviewed the recent published literature on stereotactic techniques for GH-secreting pituitary tumors with the aim to define the efficacy and potential adverse effects of each of these techniques.
GH-secreting pituitary adenoma is responsible for acromegaly, a disorder characterized by significant morbidity and mortality due to musculoskeletal, cardiovascular, metabolic, and pulmonary complications [
Surgery, radiotherapy (RT), and medical therapies, including dopamine agonists, somatostatin-receptor ligands, and the GH-receptor antagonist pegvisomant are available treatments for patients with GH-secreting pituitary adenomas. Transsphenoidal surgery is often employed in the initial management of acromegaly. Remission of disease is achieved in up to 60% of patients [
More recently, stereotactic radiation techniques have been employed in patients with GH-secreting pituitary adenomas to deliver more localized irradiation with a steeper dose gradient between the tumor and the surrounding normal tissue in order to minimize radiation-induced toxicity while improving its effectiveness [
In this paper, we present a critical analysis of the more recent available literature on fractionated stereotactic radiotherapy and radiosurgery for GH-secreting pituitary adenomas, in an attempt to define the efficacy, safety, and role of the individual stereotactic techniques.
SRS using either a multiple cobalt-60 gamma radiation-emitting sources gamma knife (GK) or a LINAC has been extensively employed in the last two decades in patients with small brain tumors [
Instead of using an array of cobalt sources, Linac SRS utilizes X-rays which are derived from colliding accelerated electrons with a target metal. The treatment is delivered with the use of multiple arcs or beams resulting in high-dose differential between the target and normal brain tissue. Isodose gradients are improved by the use of multiple isocentre plans, intensity modulation of the beams, restriction of gantry angles and arc lengths, and microcollimation. Regardless the superiority in terms of dose delivery and distribution claimed for each of these techniques, the reported clinical efficacy and toxicity are similar.
CyberKnife (Accuray, Sunnyvale, CA) is a relatively new technological device which combines a mobile linear accelerator mounted on a robotic arm with an image-guided robotic system. The system allows for frameless SRS achieving the same level of targeting precision as frame-based SRS. Patients are fixed in a thermoplastic mask, and the treatment can be delivered in form of hypofractionated regimen in patients with tumors involving the optic apparatus and who are not suitable for SRS [
Particle radiation has been also applied successfully in the treatment of pituitary adenomas. The physical properties of proton irradiation can offer superior conformality in dose distribution when compared to photons, and the advantage becomes more apparent for large volumes [
FSRT is a stereotactic technique in which a variable number of fractions are delivered to a target by a modified LINAC-based accelerator. Although FSRT uses the same planning system as SRS, patients undergoing FSRT are usually immobilized in a high-precision frameless stereotactic fixation system, including infrared camera guidance [
The decision on whether to use SRS or FSRT for pituitary tumors mainly depends on the volume of the target lesion and its proximity to sensitive structures. SRS is usually offered to patients with relatively small adenomas not in close proximity of the optic apparatus. A well-defined dose-dependent risk of radiation optic neuropathy exists following single doses of irradiation, and current practice of SRS aims to avoid irradiating the optic apparatus to single doses beyond 10 Gy [
By contrast, there is no restriction to the size of pituitary adenoma suitable for SRT when a conventional fractionation is used, since the delivered total doses are within tolerance of normal brain structures, including the optic apparatus.
Results of 29 recent published studies including 1215 patients with GH-secreting pituitary adenomas treated with SRS are showed in Table
Summary of results of recent series on SRS for GH-secreting pituitary adenomas.
Authors | patients No | Type of SRS | Total dose Gy | followup median (months) | Tumor control (%) | Hormone normalization (%) | Late toxicity (%) | |
Visual | Hypopituitarism | |||||||
Morange-Ramos et al. [ | 15 | GK SRS | 28 | 20 | NA | 20 | 5 | 16 |
Lim et al. [ | 20 | GK SRS | 25 | 26 | 92.5 | 30 | 5 | 5 |
Landolt et al. [ | 16 | GK SRS | 25 | 17 | NA | 50 | 0 | 0 |
Kim et al. [ | 11 | GK SRS | 28.7 | 27 | NA | 35 | 0 | 0 |
Inoue et al. [ | 12 | GK SRS | 21 | >24 | 94 | 58 | NA | NA |
Mokry et al. [ | 10 | GK SRS | 16 | 46 | 100 | 31 | 0 | NA |
Izawa et al. [ | 29 | GK SRS | 22.5 | >6 | 100 | 41 | 0 | 0 |
Zhang et al. [ | 68 | GK SRS | 31 | 32 | NA | 40 | 1.3 | 4 |
Landolt et al. [ | 31 | GK SRS | 25 | 19.2 | NA | 69 | NA | NA |
Ikeda et al. [ | 17 | GK SRS | 25 | 58 | NA | 82 | 0 | 0 |
Pollock et al. [ | 26 | GK SRS | 20 | 36 | 100 | 47 | 0 | 16 |
Swords et al. [ | 13 | LINAC SRS | 8–15 | 25 | 100 | 35 | 0 | 0 |
Choi et al. [ | 12 | GK SRS | 28.5 | 43 | 100 | 30 | 0 | 0 |
Attanasio et al. [ | 30 | GK SRS | 20 | 46 | 100 | 30 at 5 years | 0 | 6.7 |
Jane et al. [ | 64 | GK SRS | 15 | >18 | NA | 36 | 0 | 28 |
Castinetti et al. [ | 82 | GK SRS | 26 | 49.5* | NA | 17 | 1.2 | 17 |
Gutt et al. [ | 44 | GK SRS | 23 | 22 | 100 | 48 | NA | NA |
Kobayashi et al. [ | 67 | GK SRS | 18,9 | 63 | 100 | 17 | 11 | 15 |
Jezkov | 96 | GK SRS | 32 | 53.7 | 100 | 44 at 5 years | 0 | 27.1 |
Voges et al. [ | 64 | LINAC SRS | 16,5 | 54.3 | 97 | 14 and 33 at 3 and 5 years | 1.4 | 13 and 18 at 3 and 5 years |
Petit et al. [ | 22 | Proton SRS | 20 CGE | 75.6 | 100 | 59 | 0 | 38 |
Pollock et al. [ | 46 | GK SRS | 20 | 63 | 100 | 11 and 60 at 2 and 5 years | 0 | 33 at 5 years |
Vik-Mo et al. [ | 53 | GK SRS | 26.5 | 67 | 100 | 58 and 86 at 5 and 10 years | 3.8 | 10 at 5 years |
Jagannathan et al. [ | 95 | GK SRS | 22 | 57 | 98 | 36 and 47 at 3 and 5 years | 4 | 34 |
Losa et al. [ | 83 | GK SRS | 21,5 | 69 | 97 | 52 and 85 at 5 and 10 years | 0 | 10 at 10 years |
Ronchi et al. [ | 35 | GK SRS | 20 | 114 | 100 | 15 and 46 at 5 and 10 years | 0 | 69 |
Wan et al. [ | 103 | GK SRS | 21,4 | 67 | 95 | 37 | 0 | 6 |
Hayashi et al. [ | 25 | GK SRS | 25.2 | 36 | 100 | 40 | 0 | 0 |
Iwai et al. [ | 26 | GK SRS | 20 | 84 | 96 | 17 and 47 at 5 and 10 years | 0 | 8 |
Milker-Zabel et al. [ | 20 | FSRT | 52.2 | 61 | 100 | 80 at 5 years | 5 | 15 |
Colin et al. [ | 31** | FSRT | 50.4 | 80 | 99 | 20 and 50 at 5 and 10 years | 0 | 37 |
Minniti et al.[ | 18** | FSRT | 45 | 39 | 98 | 50 at 5 years* | 0 | 22 |
Imran et al. [ | 12 | FSRT | 50 | 28.5 | 92 | 33 | 0 | 8 |
Roug et al. [ | 34 | FSRT | 54 | 45 | 91 | 36 at 5 years | NA | 29 |
NA not assessed.
*mean followup; **acromegalic patients included in series of FSRT for either secreting or non secreting pituitary tumors.
In a retrospective analysis of 83 patients with acromegaly treated with GK SRS at University of Milan San Raffaele between 1994 and 2006, the reported actuarial biochemical remission rates were 30%, 52%, and 85% at 3, 5, and 10 years, respectively [
Remission of disease after SRS has been associated with pretreatment levels of GH and/or IGF-I levels in some series [
Marginal doses of 15–34 Gy have been employed for the treatment of GH-secreting pituitary adenomas. In the majority of studies, higher doses were not associated with higher rate of remission or faster normalization of GH/IGF-I levels. Thus, a marginal dose of about 20–25 Gy seems appropriate to achieve either tumor control or hormonal normalization.
The concomitant use of somatostatin analogs at the time of SRS as negative predictor of biochemical remission remains matter of debate. Although in Landolt et al. [
The reported overall rate of serious complications after SRS is low (Table
A recent experience of proton SRS showed a biochemical remission in 50% of 22 patients with GH-secreting pituitary adenomas, with a median time to complete response of 30.5 months [
Five studies report on the use of FSRT in 115 patients with GH-secreting pituitary adenomas [
A low-radiation-induced toxicity has been reported after FSRT, even in the case of large tumors involving the optic apparatus. Hypopituitarism has been reported in 8–37% of patients at median followup ranging from 28 to 82 months, whereas the reported incidence of optic neuropathy is 1–5%. No cases of CVA and second tumors have been reported after FSRT. Since the incidence of such complications increases with time, large series and longer followups need to demonstrate the potential clinical advantages of treating less normal brain at high doses achieved with the use of the stereotactic techniques. Similarly, because of the lack of formal cognitive function testing and quality of life assessment after FSRT, the potential superiority of stereotactic techniques as compared with 3D conformal remains to be clarified.
Initial experiences with the application of CyberKnife in treating patients with acromegaly are promising [
There is much debate about the relative efficacy of SRS and SRT. Currently reported results suggest similar results in terms of tumor control and biochemical remission of acromegaly.
A faster decline of serum GH concentration after GK SRS as compared with FSRT has been reported by some authors [
In absence of comparative studies, the choice of the radiation technique is based on tumor characteristics. SRS is usually offered to patients with relatively small adenomas less than 3 cm in size and away more than 2-3 mm from the optic chiasm. FSRT should be preferred in patients with large tumors in close proximity of optic apparatus, since the treatment is delivered within the radiation tolerance limits of cranial nerves, including the optic apparatus.
SRS and FSRT represent effective treatment modalities of irradiation for patients with persistent active GH/IGF-I hypersecretion after surgery and/or during medical therapy, providing a comparable high rates of tumor control and endocrinological remission with low morbidity. Treating less normal brain by higher radiation doses is a clear technical improvement of modern RT which translates into clinical benefit in terms of reduction of late effects of radiation. In most centres, SRS is a convenient approach for patients with relatively small residual GH-secreting tumors, while FSRT is usually reserved to patients with larger tumors not amenable to SRS. Prospective studies comparing SRS with FSRT would be of value to evaluate the long-term efficacy and toxicity of the techniques. Efficacy and toxicity of hypofractionated treatment schedules need to be explored in future studies.