Introduction

Stage IV metastatic disease will be found in approximately 30–40% of patients with non-small cell lung cancer (NSCLC). The brain represents the most common metastatic site for primary lung. The presence of a solitary brain metastasis is a common neurological situation in lung cancer patients ( ). The best treatment approach for solitary brain metastasis is still not well defined. Whether a single brain metastasis should be submitted to surgical resection or stereotactic radiosurgery (SRS), and which patients should receive whole brain adjuvant radiotherapy (WBRT), are still debated issues ( ).

SRS is an effective and non-invasive approach for oligometastatic disease to the brain, usually one to four brain metastases. The size of the lesion is another crucial factor influencing the possibility to use SRS. When a single, small brain lesion is the only site of systemic disease progression, SRS can achieve high uncomplicated local control rates (80–90% local control rates), with median survival improving from 2 to 9–12 months in patients with single or multiple metastatic brain lesions ( ). The most commonly available radiosurgery technology uses high-energy X-rays generated from a clinical linear accelerator, or “LINAC”. Here we analyze and discuss technical, radiobiological and clinical aspects of LINAC-based radiosurgery for isolated brain metastasis from NSCLC.

Technical and Technological Aspects

The term “stereotactic radiosurgery” was introduced in 1951, when Lars Leksell talked about a method for “the non-invasive destruction of intracranial lesions”. The second step was the introduction of his first stereotactic frame, developing a few years later the definitive concept of radiosurgery ( ). Thanks to the possibility of localizing, with a high grade of accuracy, a target in three-dimensional space, Leksell affirmed that narrow beams of radiation intersecting at a common point could allow neurosurgeons to concentrate high doses of radiation to a target volume in the brain (Leskell et al, 1983). With this system of delivery, lesions located deeply within the brain or at the base of the skull could be treated with minimal damage to surrounding healthy critical structures. Further research on this field by his group culminated in the development of the Gamma knife ( ). In the late 1980s, at the University of Florida, a multidisciplinary team of neurosurgeons, radiation physicists and engineers studied the development of the first LINAC-based radiosurgery system ( ). From this date, with this LINAC system, more than 2000 patients have been treated with radiosurgery at the University of Florida. From the beginning of the 1990s, several other types of LINAC-based radiosurgery systems are currently available in the commercial market and, to date, it is the most common approach for SRS delivery worldwide.

SRS is a radiation technique able to deliver, in a single fraction, an extremely high dose of ionizing radiation to an intracranial target; the lesion is usually identified using stereotactic parameters on various types of imaging, performed before and during the procedure. The radiation tolerance of normal tissue is volume-dependent, and SRS exploits this concept delivering high conformal doses to small targets ( ). With stereotactic delivery modality, the expected side effects depending on radiation dose delivered to critical structures are reduced, minimizing or eliminating the margins of normal tissue surrounding the lesion usually included in the planning treatment volume with conventional radiotherapy techniques ( ). The most commonly available SRS systems are based on high-energy X-rays generated from a clinical LINAC. The intersection of the axis of rotation of the gantry and the treatment couch and the central axis of the photon beam is a point defined as the LINAC isocenter. Planning Target Volumes could be treated using single or multiple isocenter plans. Classic stereotactic LINAC technology uses circular collimators of various diameters (from 5 mm to 50 mm) to reduce the beam penumbra, minimizing irradiation of healthy tissues surrounding the target. More recently, the level of dose conformation to the target can be further increased using micromultileaf collimators that can adapt isodoses to the shape of the target; this is certainly true and quite useful for irregularly-shaped (non-spherical) target volumes. With the intent to increase the performance to conform as better the dose to the target as possible, multiple fixed fields or dynamic-arc radiosurgery systems have been developed and applied in clinical practice. With this technology, LINAC-based SRS systems could be able to generate dose distributions with a conformity level similar to those reported for multiple isocenters Gamma knife plans ( ). With the same “philosophy” of the multiple beam Gamma knife, Cyber knife uses multiple fixed-beam positions and multiple isocenters for planning. The Cyber knife System is a LINAC-based system for SRS and stereotactic body radiotherapy: it combines a miniaturized LINAC equipped on a robotic arm with a system for target tracking with two orthogonal diagnostic X-ray generators; subsequently, the beam realignment is performed with a software for correction. Cyber knife is a 6-MeV LINAC and uses circular collimators of various diameters attached to a six-axis robotic arm. Stereotactic frames are not normally useful for target position definition.

Intensity Modulated Radiation delivery (IMRT) can also be used as an SRS treatment delivery technique, further to improve conformity indices, when compared with conventional stereotactic delivery systems. IMRT treatment must be carefully performed using three-dimensional images of the patient in conjunction with complex dose calculation algorithms to create the dose intensity pattern for the best conformity to the treatment volume shape. Usually, various combinations of several intensity-modulated fields with different beam directions generate a custom tailored radiation dose. The result is that dose distribution maximizes target coverage, while better sparing adjacent healthy tissues exposed to higher radiation doses.

Treatment positioning is crucial for precision delivery of SRS: the patient’s head is aligned to the LINAC beam and immobilized with a personalized thermoplastic mask. In modern LINAC systems dedicated to SRS, a robotic couch is provided to reduce uncertainties of patient positioning compared to planned parameters, and volumetric Image Guidance systems (IGRT) are used to improve set-up error corrections before and after treatment. To guarantee high accuracy in the SRS treatment, diagnostic three-dimensional images and highly conformal delivery are combined in the same treatment unit. Modern imaging systems integrated on modern LINAC represent a further evolution step for treatment units in terms of technological advances. Because patient position uncertainties, due to movement between image acquisition and dose delivery, is a critical reason for imprecision of the entire treatment, IGRT methods are based on non-invasive immobilization devices and patient position tracking systems. IGRT is available in the market of LINAC as various types of integrated imaging on board systems: very high quality of images, as well as KV cone beam CT or MeV CT and/or electronic portal images, are commonly available on LINACs for SRS and precise treatments. Due to the possibility to correct set up errors with a very high resolution quality of images, more comfortable frameless immobilization devices are used even more for SRS patients when combined with IMRT-IGRT delivery.

Volumetric IMRT techniques (VMAT: Volumetric Modulation Arc Therapy) and other rotational delivery systems have also significantly reduced delivery treatment time compared to the previous classical forms of IMRT delivery (static or dynamic). Flattening Filter Free (FFF) modality for LINAC is currently under evaluation as a new tool to deliver high doses combined with Volumetric Modulation Arc Therapy: it was recently shown that very high dose per fraction (more than 20 Gy) can be delivered in a few minutes with FFF beams. With FFF beams, it was possible to increase the dose/rate by a factor of four, having more dose on the central axis and less peripheral doses ( ). For a particular type of tumor diameter, it could be the perfect approach to reduce normal tissue exposure, to reduce treatment time and potentially to guarantee an improved tumoricidal intensity of the dose, due to the increased dose rate. FFF technical aspects could represent a possible revolution in SRS, allowing clinicians to use LINAC as a safe, effective and extremely rapid arm to treat brain as well as extracranial lesions. Even if promising data regarding feasibility of stereotactic radiotherapy and FFF mode have been recently published, further studies on FFF delivery are still ongoing and warranted ( ). However, definitive data of clinical results are awaited to consolidate the safety of such an approach for each type of tumor site and size.

Radiobiological Issues and Clinical Data

SRS is a powerful arm in the radiation treatment of brain tumors. The biological effectiveness of a single high radiation dose provides excellent results in terms of tumor control. Crucial conditions for treatment success are: small tumor size and appropriate location. In the case of a large lesion, the risk of healthy brain parenchyma damage and necrosis is high, regardless of dose conformation due to the steep dose gradient. In the case of specific critical organs, such as optic nerves or brainstem, close to or inside the planning target volumes, the technical ability to deliver an effective dose to the tumor is clearly limited by normal tissue tolerance of such specific normal structures. Thus, to optimize the clinical use of SRS, radiobiologic knowledge is essential fully to understand the potential tumoricidal effect of high single doses and to avoid potential side effects on normal tissues.

From its first applications, stereotactic radiotherapy, especially with Gamma kinfe, was used in a single fraction by clinicians. Thus, its powerful potential was strictly correlated to its ability to irradiate brain lesions with excellent dose distributions. From a radiobiologic point of view, it is recognized that a single-dose fraction could be unable to sterilize even a small tumor that contains hypoxic malignant cells. Based on the “hypoxic” rationale, a single-fraction delivered in radiosurgery is considered a suboptimal modality in the case of malignant tumors ( ). Thus, a multifractionated stereotactic radiotherapy schedule for a large intracranial malignancy is better advocated to treat this kind of lesion. On the other hand,, even for malignancies, in the case of a small treatment volume that contains little functioning brain tissue, the need for fractionation may not apply ( ). Moreover, one of the potential advantages of radiosurgery, compared to multiple sessions, is that it may be able to overcome the radioresistance of a hypoxic tumor with a radiologic ablative effect when very high doses are delivered ( ).

From a radiobiologic point of view, brain metastases are considered a category of target with early-responding tissue surrounded by late-responding normal tissue. Starting from the consideration that tumor cells are all inside the target, a high therapeutic index, based on the steep dose gradient performed by SRS, would be expected using this technique. Unlike surgery, SRS has also the potential to sterilize cells that may not be contained within the high-dose region ( ).

Reviewing the past literature on solitary brain metastases in NSCLC patients, no clear consensus on the treatment of choice was found ( ). In the case of single brain metastasis from NSCLC, the life expectancy is mainly influenced by the general condition of the patient and the definitive treatment of the primary lung tumor ( ). Prognostic factors, besides specific local considerations (site and size of the brain tumor, number of brain metastases), can guide clinicians towards a more accurate treatment decision between surgery, WBRT and SRS ( ). Although the presence of brain metastases is often considered as an end-stage of disease, the majority of patients with controlled intracranial metastases will die from systemic disease rather than from recurrence and progression of these metastases.