Ervin B. Podgorsak

Verification of segmented beam delivery using a commercial electronic portal imaging device

In modern radiotherapy, three-dimensional conformal dose distributions are achieved through the delivery of beam ports having precalculated planar distributions of photon beam intensity. Although sophisticated means to calculate and deliver these spatially modulated beams have been developed, means to verify their actual delivery are relatively cumbersome, making equipment and treatment quality assurance difficult to enforce. An electronic portal imaging device of the scanning liquid ionization chamber type yields images which, once calibrated from a previously determined calibration curve, provide highly precise planar maps of the incident dose rate. For verification of an intensity-modulated beam delivered in the segmented approach with a multileaf collimator, a portal image is acquired for each subfield of the leaf sequence. Subsequent to their calibration, the images are multiplied by their respective associated monitor unit settings, and summed to produce a planar dose distribution at the measurement depth in phantom. The excellent agreement of our portal imager measurements with calculations of our treatment planning system and measurements with a one-dimensional beam profiler attests to the usefulness of this method for the planar verification of intensity-modulated fields produced in the segmented approach on a computerized linear accelerator equipped with a multileaf collimator.

Interstitial pneumonitis following total body irradiation for bone marrow transplantation using two different dose rates

A total of 22 patients with leukemia (10 ALL, 11 AML, 1 CML) have undergone allogeneic bone marrow transplantation (BMT) by the Quebec Co-operative Group for Marrow Transplantation from 1980 to 1982. All patients received 900 cGy total body irradiation (TBI), in a single fraction, on the day preceding BMT. The first 11 patients were treated on a cobalt unit at a constant dose rate of 4.7 to 6.3 cGy/min. Six of these patients developed interstitial pneumonitis (IP). The clinical course of three patients, two with idiopathic and one with drug-induced pneumonitis, was mild and recovery was complete in all. The other three patients developed severe infectious IP and two died. The next 11 patients were treated with a sweeping beam technique on a 4 MV linear accelerator delivering a total tumor dose of 900 cGy at an average dose rate of 6.0 to 6.5 cGy/min but an instantaneous dose rate of 21.0 to 23.5 cGy/min. Eight patients developed severe IP. Five of these were idiopathic and four died. Three were infectious and all died. The fatality of interstitial pneumonitis appeared to be greater in the group treated with the sweeping beam technique.

An electronic portal imaging device as a physics tool

An electronic portal imaging device (EPID) can be used not only to acquire megavoltage patient images but also to measure certain radiation beam parameters of the linear accelerator. EPID images can be used to verify field junctions, center of collimator rotation, or radiation vs. light field coincidence. If the EPID images are calibrated in terms of dose rate, an EPID can be applied to beam penumbra measurement, collimator transmission determination, or compensator verification. Beam parameters measured with EPIDs are in close agreement with those measured with film or ionization chamber, making EPIDs reliable physics tools for quality control of various beam parameters in radiotherapy.

Determination of saturation charge and collection efficiency for ionization chambers in continuous beams

The procedure recommended by radiation dosimetry protocols for determining the collection efficiency f of an ionization chamber assumes the predominance of general recombination and ignores other charge loss mechanisms such as initial recombination and ionic diffusion. For continuous radiation beams, general recombination theory predicts that f can be determined from a linear relationship between 1/Q and 1/V2 in the near saturation region (f > 0.7), where Q is the measured charge and V the applied chamber potential. Measurements with Farmer-type cylindrical ionization chambers exposed to cobalt-60 gamma rays reveal that the assumed linear relationship between 1/Q and 1/V2 breaks down in the extreme near-saturation region (f > 0.99) where Q increases with V at a rate exceeding the predictions of general recombination theory. A comprehensive model is developed to describe the saturation characteristics of ionization chambers. The model accounts for dosimetric charge loss (initial recombination, ionic diffusion, and general recombination) and nondosimetric charge multiplication in an ionization chamber, and suggests that charge multiplication plays a significant role under typical chamber operating conditions (300 V) used in radiation dosimetry. Through exclusion of charge multiplication from the measured chamber signal Q, the model predicts the breakdown of the 1/Q vs 1/V2 relationship and shows that the final approach to saturation is governed by initial recombination and ionic diffusion which are characterized by a linear relationship between 1/Q and 1/V. Collection efficiencies calculated with this model differ by up to 0.4% from those determined through a rigorous application of general recombination theory alone.

Calibration of photon and electron beams with an extrapolation chamber

A variable air-volume, parallel-plate extrapolation chamber forming an integral part of a Solid Water phantom was built to determine the absorbed dose in Solid Water directly. The sensitive air-volume of the extrapolation chamber is controlled through the movement of the chamber piston by means of a micrometer mounted to the phantom body. The relative displacement of the piston is monitored by a calibrated mechanical distance travel indicator with a precision on the order of 0.002 mm. Irradiations were carried out with cobalt-60 gamma rays, x-ray beams ranging from 4 to 18 MV, and electron beams between 6 and 22 MeV. The absorbed dose at a given depth in Solid Water is proportional to the ionization gradient measured in the Bragg-Gray cavity region with an extrapolation chamber embedded in the Solid Water phantom. The discrepancies between the doses determined in Solid Water with our uncalibrated extrapolation chamber and doses obtained with a calibrated standard thimble ionization chamber are at most 1% for photon and electron beams at all megavoltage clinical energies. Uncalibrated extrapolation chamber thus offer a simple and practical alternative to other techniques used in output measurements of megavoltage photon and electron machines.

Hypofractionated stereotactic radiotherapy for low grade glioma at McGill University: long-term follow-up.

Small, well-defined, unresectable low-grade gliomas are attractive targets for stereotactic irradiation. Fractionated stereotactic irradiation of these targets has the theoretical benefit of increased normal tissue sparing beyond that provided by the physical characteristics of stereotactic radiosurgery. From July 1987 to November 1992, 21 patients were treated for low-grade glioma at our institution using a hypofractionated regimen of stereotactic radiotherapy. All patients had well-circumscribed, < 40 mm tumors. No patient had had prior radiotherapy. All lesions were histologically proven WHO grade I or II glial tumors. Lesions involved sensitive brain structures and were deemed unresectable. A typical dose of 42 Gy was delivered in 6 fractions over a two-week period using rigid immobilization and a linac-based dynamic stereotactic radiosurgical technique. Patients had a median age of 23 years (9–74) and were predominantly female (60%). Median tumor diameter was 20 mm. With a median follow-up for living patients of 13.3 years, the actuarial 5, 10, and 15-year overall survival rates are 76%, 71%, and 63%, respectively. Treatment was acutely well tolerated although three patients experienced late post-therapy complications. Our results and those of 241 patients treated in nine other institutional series are reviewed. Despite some examples of favorable short-term outcomes, all reported series are highly selected and thus likely biased. The data regarding the use of SRS is limited and, in our opinion, insufficient to claim a clear therapeutic advantage to SRS in the initial management of low-grade glioma. Our own results with hypofractionated stereotactic radiotherapy are similar to those expected with standard therapy.

Setup verification in linac-based radiosurgery.

A semi-automatic technique for the direct setup alignment of radiosurgical circular fields from an isocentric linac to treatment room laser cross-hairs is described. Alignment is achieved by acquiring images of the treatment room positioning laser cross-hairs superimposed on the radiosurgical circular field image. An alignment algorithm calculates the center of the radiosurgical field image as well as the intersection of the laser cross-hairs. This determines any alignment deviations and the information is then used to translate the radiosurgical collimator to its correct aligned position. Two detectors, each being sensitive to the lasers and ionizing radiation, were used to acquire the radiation/laser images. The first detector consists of a 0.3-mm-thick layer of photoconducting a-Se deposited on a 1.5-mm-thick copper plate and the second is film. The algorithm and detector system can detect deviations with a precision of approximately 0.04 mm. A device with gyroscopic degrees of freedom was built in order to firmly hold the detector at any orientation perpendicular to the radiosurgical beam axis. This device was used in conjunction with our alignment algorithm to quantify the isocentric sphere relative to the treatment room lasers over all gantry and couch angles used in dynamic stereotactic radiosurgery.

Viability of an isocentric cobalt-60 teletherapy unit for stereotactic radiosurgery.

The potential for radiosurgery with an isocentric teletherapy cobalt unit was evaluated in three areas: (1) the physical properties of radiosurgical beams, (2) the quality of radiosurgical dose distributions obtained with four to ten noncoplanar converging arcs, and (3) the accuracy with which the radiosurgical dose can be delivered. In each of these areas the cobalt unit provides a viable alternative to an isocentric linear accelerator (linac) as a radiation source for radiosurgery. A 10 MV x-ray beam from a linac used for radiosurgery served as a standard for comparison. The difference between the 80%-20% penumbras of stationary radiosurgical fields in the nominal diameter range from 10 to 40 mm of the cobalt-60 and 10 MV photon beams is remarkably small, with the cobalt-60 beam penumbras, on average, only about 0.7 mm larger than those of the linac beam. Differences between the cobalt-60 and 10 MV radiosurgical treatment plans in terms of dose homogeneity within the target volume, conformity of the prescribed isodose volume to the target volume, and dose falloffs outside the target volume are also minimal, and therefore of essentially no clinical significance. Moreover, measured isodose distributions for a radiosurgical procedure on our Theratron T-780 cobalt unit agreed with calculated distributions to within the +/- 1 mm spatial and +/- 5% numerical dose tolerances, which are generally specified for radiosurgery. The viability of isocentric cobalt units for radiosurgery will be of particular interest to centers in developing countries where cobalt units, because of their relatively low costs, provide the only megavoltage source of radiation for radiotherapy, and could easily and inexpensively be modified for radiosurgery. Of course, the quality assurance protocols and mechanical condition of a particular teletherapy cobalt unit must meet stringent requirements before the use of the unit for radiosurgery can be advocated.

Physical aspects of the angle-β concept in electron arc therapy

A technique for the determination of treatment parameters that are required to achieve a desired depth dose distribution in electron arc therapy is discussed and a method for calculating isodose distributions is presented. Both the treatment technique and the dose calculation method rely on the angle beta concept, which uniquely describes the dependence of the radial percentage depth doses in electron arc therapy on the nominal field width, isocenter depth, and virtual source-axis distance. The angle beta concept is discussed in detail and the electron pseudo-arc therapy technique used at McGill is described. Also presented is the method used to achieve dose homogeneity in target volumes treated with the pseudo-arc technique.

Calculation of surface dose in rotational total skin electron irradiation

A single-field rotational total skin electron irradiation technique has recently been developed at the McGill University for treatment of skin malignancies. The dose received by a given surface point during rotation in a uniform large electron field depends on the radius of rotation of the surface point, on the local radius of curvature of the contour in the vicinity of the point of interest, and on the shadows cast by limbs (arms upon trunk or head and neck, and legs upon each other). A method for calculating the surface dose distribution on a patient is presented accounting for the various parameters affecting the dose. A series of measurements were performed with polystyrene and a humanoid phantom, and an excellent agreement between measured and calculated dose distributions was obtained.