Pituitary adenomas are very common lesions and represent between 10% and 20% of all primary brain tumors. Autopsy and imaging studies have demonstrated that nearly 20% of the general population has a pituitary adenoma. Pituitary adenomas are broadly classified into two groups. The first category of tumor secretes an excessive amount of pituitary hormone or hormones and, consequently, causes a variety of clinical syndromes depending upon hormone(s) secreted. The most common of these is the prolactinoma, which causes amenorrhea-galactorrhea and infertility in women and impotence and infertility in men. Fortunately, prolactinomas can usually be managed medically with dopamine-agonist drugs. The second most common type of functioning pituitary adenoma produces growth hormone, causing acromegaly in adults and gigantism in children before closure of the epiphyseal plates. Corticotrophin (ACTH)-secreting tumors cause Cushing’s disease or, if bilateral adrenalectomies have been performed, Nelson’s syndrome.
The second category of pituitary adenomas consists of tumors that do not secrete any known biologically active pituitary hormones but are positive on immunostaining— most commonly for luteinizing hormone (LH), follicle-stimulating hormone (FSH), and α-subunit. These represent approximately 30% of all pituitary tumors. These so-called nonfunctioning or null cell pituitary adenomas enlarge progressively in the pituitary fossa and often extend outside the confines of the sella. Nonsecretory adenomas may cause symptoms related to mass effect whereby the optic nerves and chiasm are compressed, and a bitemporal visual field loss characteristically results. Patients with a nonsecretory adenoma often have hypopituitarism as a result of compression of the normal pituitary gland.
For all types of pituitary adenomas, recurrence as a result of tumor invasion into surrounding structures (e.g., the dura or cavernous sinus) or incomplete tumor resection is quite common. Long-term tumor control rates after microsurgery alone vary from 50% to 80%. Radiosurgery can be administered postoperatively as adjuvant therapy to inhibit growth of a residual tumor or later when clinical symptoms, laboratory studies, or imaging studies indicate recurrence. The presence of residual tumor is not uncommon in adenomas with either a suprasellar component or cavernous sinus invasion; the incidence of recurrence has been shown to correlate with dural invasion by the pituitary tumor.
In 1951, stereotactic radiosurgery was described by Lars Leksell as the “closed skull destruction of an intracranial target using ionizing radiation. In 1968, Leksell treated the first pituitary adenoma patient with the Gamma Knife. Since that time, stereotactic radiosurgery has been used to treat more than 20,000 patients with a pituitary adenoma. At the same time, great attention and effort in the field of stereotactic radiosurgery have been placed on the preservation of surrounding neuronal and vascular structures.
Radiosurgery is performed using the Gamma Knife, a linear-accelerator (LINAC) based system, or proton beams produced by a cyclotron. Gamma Knife surgery usually involves multiple isocenters of different beam diameters to achieve a dose plan that conforms to the irregular three-dimensional volumes of most mass lesions. The total number of isocenters varies depending upon the size and shape of the lesion. The recent version of the Gamma Knife (Model C) combines advances in dose planning with robotic engineering and obviates the need to set coordinates manually for each isocenter. The soon to be released Gamma Knife 4C integrates additional neuroimaging modalities (e.g., SPECT and PET) into the planning software.
In LINAC-based radiosurgery, multiple radiation arcs are employed to crossfire photon beams at a target defined in stereotactic space. Most of the currently used systems use nondynamic techniques in which the patient couch is set at an angle and the arc is moved around its radius to deliver radiation that enters the skull through many different points. Numerous techniques have been developed to enhance conformality of dose planning and delivery using LINAC-based systems. These include beam shaping and intensity modulation. Newer developments include the introduction of minileaf and microleaf collimators. The conformal beam can be delivered with the micro-multileaf collimator or conformal blocks.
Proton beam radiosurgery offers a theoretical advantage because of the quantum wave properties of protons to reduce dose delivered to tissue surrounding the target. In practice, this advantage has not been rigorously demonstrated. Moreover, a cyclotron, which is required to produce a proton beam, is only available at a limited number of centers because of financial and logistical constraints.
In preparation for any type of radiosurgery, many centers have recommended a temporary cessation of medications that suppress pituitary hormone hypersecretion (e.g., dopamine-agonists and somatostatin analogs) at the time of treatment. In 2000, Landolt et al first reported a significantly lower hormone remission rate in acromegalic patients who were receiving a somatostatin analog at the time of radiosurgery. Since then, this same group and others have documented a counterproductive effect of antisecretory medications on the rate of hormonal normalization following radiosurgery. The mechanism by which antisecretory medications lower hormonal normalization rates is unknown, but Landolt et al have hypothesized that these drugs alter cell cycling and thus potentially decrease tumor cell radiosensitivity. Moreover, the optimal time period to hold antisecretory medications in conjunction with stereotactic radiosurgery is not clear. Landolt and Lomax recommend that dopamine agonists be withheld 2 months before radiosurgery. For acromegalics, they recommend reducing antisecretory medication administration as early as 4 months before radiosurgery and completely discontinuing all antisecretory medications 2 weeks before radiosurgery. Although many centers including ours have incorporated such methodology into their treatment regimen, the radiosurgical team must weigh the potential risk and benefits of altering antisecretory medication administration.
The effective delivery of radiation to a target requires clear and accurate imaging of that target. Over the past 20 years, significant advances in neuroimaging have increased the efficacy and precision of radiosurgical treatment of pituitary lesions. Tumor localization for dose planning is better achieved with enhanced coronal magnetic resonance (MR) than with computed tomography (CT) imaging. An MRI sequence consisting of postcontrast, thin-slice (e.g., 1 mm) volume acquisition is typically used to define the tumor. In patients with previous surgery, fat suppression techniques can prove useful for differentiating tumor from surgical fat grafts. In the pre-MRI era, CT was used routinely. However, now it is generally reserved for patients who cannot undergo an MRI (e.g., a patient with a pacemaker). For hormonally active lesions, if there is no obvious tumor, radiosurgery may still be successful in achieving hormonal normalization. In this case, the entire sellar region (including the inferior dura) is selected as the radiosurgical target when no definitive tumor is visible.
As part of the preoperative evaluation, we routinely perform thorough endocrinological testing and a neuro-ophthalmological evaluation. Frame placement at the University of Virginia is done in the main operating room with standard sterile technique. Monitored anesthesia is performed by a neuroanesthesiologist, and both intravenous and local anesthetic are typically administered. This surgical protocol affords optimal frame placement and a pain-free experience for the patient. For pediatric patients, the entire procedure including frame placement, neuroimaging acquisition, and gamma surgery are often performed under general anesthetic.
After frame placement and stereotactic image acquisition, dose planning is performed. Through the strategic selection of isocenters, gamma angle, prescription dose, beam blocking patterns, and isodose selection, the borders of the tumor can be encompassed and a suitable radiation dose delivered. One must take into account the radiation fall-off characteristics unique to the type of unit used. If fractionated radiation therapy has previously been administered, the dose and timing of that treatment must be considered when selecting the dose for radiosurgery.
Radiosurgical Goals and Expectations Based upon the Literature
For patients with pituitary adenomas, radiosurgery is meant to inactivate the tumor cells thereby preventing tumor growth and, for functioning adenomas, lowering hormonal overproduction to normal in addition to preventing tumor growth. Ideally, these goals are met without damaging the normal pituitary gland and surrounding vascular and neuronal structures. Moreover, radiosurgery should be carried out in such a way as to avoid delayed, radiation-associated secondary tumor formation.
A total of 35 peer-reviewed studies including 1621 patients were reviewed. *
* , , , , .Results of these studies are summarized in Tables 28-1 through 28-5 .
|Authors (Year)||Radiosurgery Unit||No. of Patients||Mean or Median Follow-up||Margin Dose||Growth Control|
|Martinez et al (1998)||GK||14||36||16||100|
|Lim et al (1998)||GK||22||26||25||92|
|Mitsumori et al (1998)||LINAC||7||47||15||100|
|Witt et al (1998)||GK||24||32||19||94|
|Yoon et al (1998)||LINAC||8||49||17||96|
|Hayahsi et al (1999)||GK||18||16||20||92|
|Mokry et al (1999)||GK||31||21||14||98|
|Izawa et al (2000)||GK||23||28||22||94|
|Shin et al (2000)||GK||3||19||16||100|
|Feigl et al (2002)||GK||61||55||15||94|
|Sheehan et al (2002)||GK||42||31||16||98|
|Wowra and Stummer (2002)||GK||30||58||16||93|
|Petrovich et al (2003)||GK||56||41||15||100|
|Muramatsu et al (2003)||LINAC||8||30||15||100|
|Pollock et al (2003)||GK||33||43||16||97|
|Losa et al (2004)||GK||54||41||17||96|
|Authors (Year)||Radiosurgery Unit||No. of Patients||Mean or Median Follow-up||Margin Dose||Endocrine Cure|
|Levy et al (1991)||Proton/helium beam||64||NR||NR||86|
|Ganz et al (1993)||GK||4||18||25||50|
|Martinez et al (1998)||GK||3||36||24||100|
|Lim et al (1998)||GK||4||26||25||25|
|Mitsumori et al (1998)||LINAC||5||47||15||40|
|Morange-Ramos et al (1998)||GK||6||20||28||67|
|Witt et al (1998)||GK||25||32||19||28|
|Yoon et al (1998)||LINAC||1||49||17||NR|
|Hayashi et al (1999)||GK||10||16||24||10|
|SH Kim et al (1999)||GK||8||27||29||62|
|Laws et al (1999)||GK||50||NR||NR||58|
|Mokry et al (1999)||GK||5||56||17||33|
|Izawa et al (2000)||GK||12||28||22||17|
|Sheehan et al (2000)||GK||43||44||20||63|
|Shin et al (2000)||GK||7||88||32||50|
|Hoybye et al (2001)||GK||18||204||NR||83|
|Feigl et al (2002)||GK||4||55||15||NR|
|Kobayashi et al (2002)||GK||20||64||29||35|
|Pollock, Nippoldt et al (2002)||GK||9||42||20||78|
|Petrovich et al (2003)||GK||4||41||15||50|
|Choi et al (2003)||GK||5||42.5||28.5||56|
|Authors (Year)||Radiosurgery Unit||No. of Patients||Mean or Median Follow-up||Margin Dose||Endocrine Cure|
|Ganz et al (1993)||GK||4||18||19.5||25|
|Martinez et al (1998)||GK||7||36||25||71|
|Landolt et al (1998)||GK||16||NR||25||81|
|Lim et al (1998)||GK||20||26||25||38|
|Mitsumori et al (1998)||LINAC||1||47||15||0|
|Morange-Ramos et al (1998)||GK||15||20||28||20|
|Witt et al (1998)||GK||20||32||19||20|
|Yoon et al (1998)||LINAC||2||49||17||50|
|Hayashi et al (1999)||GK||22||16||24||41|
|MS Kim et al (1999)||GK||2||12||22||0|
|SH Kim et al (1999)||GK||11||27||29||46|
|Laws et al (1999)||GK||56||NR||NR||25|
|Mokry et al (1999)||GK||16||46||16||31|
|Izawa et al (2000)||GK||29||28||22||41|
|Shin et al (2000)||GK||6||43||34||67|
|Zhang et al (2000)||GK||68||34||31||96|
|Fukuoka et al (2001)||GK||9||42||20||50|
|Ikeda et al (2001)||GK||17||48||25||82|
|Feigl et al (2002)||GK||9||55||15||NR|
|Pollock, Nippoldt, et al (2002)||GK||26||42||20||42|
|Attanasio et al (2003)||GK||30||46||20||37|
|Petrovich et al (2003)||GK||6||41||15||100|
|Muramatsu et al (2003)||LINAC||4||30||15||50|
|Choi et al (2003)||GK||12||42.5||28.5||50|