UNIVERSITY OF WISCONSIN-LA CROSSE Graduate Studies USE OF STEREOTACTIC BODY RADIATION THERAPY FOR INOPERABLE NON- SMALL CELL LUNG TUMORS

Transcription

1 UNIVERSITY OF WISCONSIN-LA CROSSE Graduate Studies USE OF STEREOTACTIC BODY RADIATION THERAPY FOR INOPERABLE NON- SMALL CELL LUNG TUMORS A Research Project Report Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science in Medical Dosimetry Kevan B. Grimm College of Science & Health Medical Dosimetry Program May 2013

2 2 USE OF STEREOTACTIC BODY RADIATION THERAPY FOR INOPERABLE NON- SMALL CELL LUNG TUMORS By Kevan B. Grimm We recommend acceptance of this project report in partial fulfillment of the candidate's requirements for the degree of Master of Science in Medical Dosimetry The candidate has met all of the project completion requirements. Nishele Lenards, M.S. Graduate Program Director May 6, 2013 Date

3 3 The Graduate School University of Wisconsin-La Crosse La Crosse, WI Author: Grimm, Kevan B. Title: Use of Stereotactic Body Radiation Therapy for Inoperable Non-Small Cell Lung Tumors Graduate Degree/ Major: MS Medical Dosimetry Research Advisor: Nishele Lenards, M.S. Month/Year: May 2013 Number of Pages: 49 Style Manual Used: AMA, 10 th edition Abstract Treatment options for patients who present with inoperable stage I and II (T1-2, N0, M0) non-small cell lung cancer (NSCLC) consist of surgery, external beam radiation therapy (EBRT), or observation. Because observation has such poor overall survival (OS) rates and surgery is limited to only certain individuals, EBRT presents itself as viable option to patients. Conventional radiation therapy has local recurrence rates and 5-year OS rates that are drastically worse than surgical resection. In an effort to reach a more favorable outcome, attempts have been made to increase radiation doses to the tumor while maintaining safe levels of radiation to surrounding tissues. With recent technological advancements, the use of stereotactic body radiation therapy (SBRT) has become a viable treatment option to increase the OS of NSCLC patients. This study will compare conventional radiation treatments to stereotactic body radiation therapy treatments using dosimetric values and their relationship to treatment outcomes.

4 4 The Graduate School University of Wisconsin - La Crosse La Crosse, WI Acknowledgments I would like to thank the Radiation Oncology departments at Akron City Hospital and McLeod Health for all of their effort in assisting me with this research.

5 5 Table of Contents... Page Abstract...3 List of Tables...6 List of Figures...7 Chapter I: Introduction...8 Statement of the Problem...12 Purpose of the Study...12 Assumptions of the Study...12 Definition of Terms...13 Limitations of the Study.15 Methodology...15 Chapter II: Literature Review...16 Chapter III: Methodology...26 Sample Selection and Description...26 Instrumentation...26 Data Collection Procedures...26 Data Analysis...27 Limitations...27 Summary...27 Chapter IV: Results...29 Item Analysis...29 Chapter V: Discussion...32 Limitations...32 Conclusions...33 Recommendations...34 References...48

6 6 List of Tables Table 1: Tumor (T) descriptors of the TNM staging system...35 Table 2: Nodes (N) descriptors of the TNM staging system...36 Table 3: Metastasis (M) descriptors of the TNM staging system...36 Table 4: Lung tumor staging...37 Table 5: Lung toxicity...37

7 7 List of Figures Figure 1: R Figure 2: D 2cm...39 Figure 3: PTV Figure 4: PTV Figure 5: MLD...42 Figure 6: Lung V Figure 7: Mean Dose- Spinal Cord...44 Figure 8: Maximum Dose-Spinal Cord...45 Figure 9: Mean Dose-Heart...46 Figure 10: Maximum Dose-Heart...47

8 8 Chapter I: Introduction Non-small cell lung cancer (NSCLC) continues to be the leading cause of cancer death with over one million cases presented yearly in the United States alone. 1,2 Approximately 75% of all lung cancers are NSCLC and the fraction of patients presenting with stage I and II (T1-2, N0, M0) NSCLC is rapidly growing. Some attribute this to an increase in early detection due to the use of computed tomography (CT) scanners for common tests. 1,2 The rise in use of CT scans for routine exams has improved the chance that the disease will be diagnosed at an earlier stage. For patients that present with stage I and II (T1-2, N0, M0) NSCLC, the gold standard for treatment is a surgical resection that will include either a lobectomy or pneumonectomy with nodal resection. 2 Surgical resection results in a 5-year overall survival (OS) rate of 60 to 70%. 3 However, not all patients who present with early stage NSCLC are eligible for surgery. Many conditions such as forced expiratory volume, reduced diffusion capacity, hypoxemia and/or hypercapnia are reasons an individual may be deemed inoperable, forcing them to look to other options for treatment. 3 Once a patient is considered inoperable, their choices of treatments are to have a wedge resection, (a less invasive surgery, with a very poor outcome), external beam radiation therapy (EBRT), or observation. 1 Wedge resection is ineffective in local disease control and may actually decrease OS and observation alone can decrease 5- year survival rates to less than 10%. 1,2 In order for a patient to even be considered for radiation therapy (RT), they must have been deemed inoperable. This is due to the decreased local control rates and OS of RT compared to surgery. 2 Patients may also refuse surgical resection for personal reasons and be recommended for RT; however, this decision may prevent them from being eligible for a clinical trial. In many cases, EBRT is the most viable option for patients who cannot undergo surgery. Once the patient has selected EBRT as their treatment option, a few other things must be considered. According to Timmerman et al 3 tumor size and histology may both be limiting factors. Lesions staged T1 or T2 must be under 7cm to meet the requirements for SBRT. 3,4 In determining the eligibility of treatment due to the histology of the tumor, the histology can be determined by either a biopsy or cytology. The cell types eligible for SBRT are squamous cell carcinoma, adenocarcinoma, large-cell carcinoma, bronchioloalveolar cell carcinoma, or NSCLC (not otherwise specified). 3 In cases with no known histological diagnosis a Positron Emission Tomography (PET)/CT scan may be adequate. 5 Tumor location may also affect a patient s eligibility. Non-small cell lung cancer tumors are typically either peripheral (> 2cm from any

9 9 mediastinal structures) or central (< 2cm from any mediastinal structures). 3 Physicians may be hesitant to treat central lesions due to important critical structures, such as the spinal cord and heart, which may be placed at risk during radiation treatment. While EBRT is considered a viable option, the expected 5-year OS rate is 10% -30% with local recurrence occurring in 55% -70% of cases. 3 In these situations a standard EBRT dose of 45 Gy to 66 Gy is given over a 5-7 week time period. 6 The poor local control and OS of EBRT has led to attempts to increase the effectiveness of the treatment. The Radiation Therapy Oncology Group (RTOG) conducted several trials dating back to the 1980s attempting to increase dose to tumors for NSCLC. 7 These trials had questionable results, showing a slight increase in OS with the first levels of dose escalation, which at that time was 69.6 Gy for standard fractionation. Unfortunately, as the dose was escalated to 74.4 Gy and 79.2 Gy, the OS dropped dramatically. 7 Many professionals believe this was a result of increased lung toxicity and morbidity to surrounding critical structures as the dose was escalated. The invention of Intensity Modulated Radiation Therapy (IMRT) improved the dose conformality that could be achieved on a linear accelerator. Intensity Modulated Radiation Therapy was commonly used by radiation oncology facilities by the late 1990s. This new technology was able to increase dose conformality while reducing the lung toxicity observed during earlier attempts to increase dose. The effect of radiation on cells in the human body is measured in biological equivalent dose (BED); as the BED increases, cell destruction increases. A dose of 60 Gy in 30 fractions has a BED of 72 Gy. In comparison, the same dose of 60 Gy delivered in only 5 fractions results in a BED of 132 Gy. The goal of Stereotactic Body Radiation Therapy (SBRT) is ablation, or death, of the tumor while sparing the surrounding structures. Reaching a high dose is necessary for tumor ablation, but for doses outside the target, the increased BED equates to increased damage. Every organ in the human body has a tolerance dose (TD). An organ s TD is the maximum amount of radiation the organ can receive. 4 Any dose delivered beyond an organ s TD may potentially cause a patient to encounter adverse side effects. For example, in cases such as treating near the spinal cord, too much radiation can result in paralysis and even death. If a patient has had previous RT where an organ in the area received dose, that dose must be subtracted from the tolerance to ensure the TD is not surpassed. As previously mentioned, dose fractionation alters the BED of radiation and if the previous treatment being evaluated differed in fractionation from the current SBRT fractionation an equation must be used to calculate the correct total dose that will be received by that organ. 4 In some cases not being treated outweighs

10 10 the side effects. Tolerance dose are usually listed in relationship to standard fractionation. Therefore, to compare doses and adequately evaluate an SBRT treatment, the doses must be scaled. Failure to make this adjustment can cause injury or even death of the patient. The standard fractionation for an inoperable lung lesion is 45 Gy-66 Gy divided into treatments of 1.8 Gy-2 Gy per day. Typical radiation scheduling is Monday through Friday. This schedule was developed after many years of trial and error in order to allow adequate cell repair during the time between treatments and the two-day break over weekends. Conventional radiation treatment for lung cancer lasts 6-7 weeks. Current SBRT dose schemas range from Gy over 3-5 fractions of 12 Gy-20 Gy per fraction. 1,3,7 Typically 1-2 days is required between fractions due to additional time needed for cell recovery with the larger doses per fraction. Altering the fractionation schedule for SBRT treatment can affect critical structure doses, dose conformality, the BED, and the overall treatment outcome. Early attempts to hypofractionate treatments for NSCLC failed due to inadequacies in the treatment technologies at the time and the use of kilo-electron volts (KeV) equipment which did not penetrate as deep as the mega-electron volts (MeV) used today. This resulted in excessive doses to the superficial regions of the body. 4 In the 1990s, hypofractionation was used with the introduction of stereotactic radiosurgery (SRS) as a technique that allowed for the delivery of large doses in one fraction to an intracranial lesion. Stereotactic radiosurgery showed that high doses of radiation could be focused on a small amount of tissue while maintaining tolerable doses to the surrounding normal tissues. Stereotactic radiosurgery was achievable with intracranial lesions due to superior immobilization of the head and neck, the utilization of a grid system, which assisted in tumor localization, and the dose conformality of the machines being used for SRS. With the success of SRS for intracranial lesions, physicians began to explore its use on extracranial lesions. Treating tumors located in other parts of the body would also require the adaptation of patient immobilization, target localization, and dose conformality similar to the brain SRS treatments. Using SRS on lung lesions did not have a favorable outcome and a decision to use multiple fractions was made. Multiple fraction schemes led to the development of SBRT. Stereotactic Body Radiation Therapy utilizes the same techniques as SRS, but delivers the dose over 3-10 fractions. Stereotactic Body Radiation Therapy is a treatment that can achieve tumor ablation; however, if not used correctly and with the proper immobilization and localization techniques, patients can develop terrible toxicities. 8

11 11 External beam radiation therapy has numerous variables that could increase the chance of missing the target. Factors that must be considered prior to radiation treatments include patient motion (both voluntary and involuntary), set-up error (human error by radiation therapist when preparing the patient for treatment), equipment limitations, and dose build up regions. The planning process begins by creating target volumes. The physician will outline the gross extent of the tumor on a CT scan and that is called the gross tumor volume (GTV). 4 A margin is added to the GTV to account for microscopic disease and the dose build up region, which is known as the clinical target volume (CTV). Tumors that have a large range of motion, such as those in the lungs, may require an internal target volume (ITV), which can be created by one of several methods. The current trials suggest the use of four-dimensional (4D) scans, which take into account the patients breathing cycle. However, many facilities do not have 4D capabilities and struggle with its integration into treatment planning. 5 Many facilities suppress the patient s breathing by using a restricting band wrapped around the patient s diaphragm reducing movement to the point in which the breathing motion becomes a minimal issue. 5 Respiratory gating can also be used in conjunction with the 4D scans. Respiratory gating gives the ability to account for breathing motion by turning the radiation beam on and off when the tumor moves in and out of the field. Other facilities will initially acquire 3 CT scans, one of normal breathing and then one each with the patient holding inspiration and expiration. The ITV is created by contouring several GTVs from CT scans taken at various phases of the respiratory cycle to best represent full tumor motion. From the ITV, a third margin is added to account for set-up errors and patient motion, which is known as the planning target volume (PTV). The size of these additional margins may vary depending on the type of treatment the patient will be receiving. For some treatments the patient motion and set-up error may vary. When utilizing an ITV the final margins are decreased because the ITV accounts for breathing. Patient immobilization and tumor localization techniques must be improved to decrease the possibility of missing the target and to allow smaller margins to be used for planning. Missing the target by more than a few millimeters will result in both increased dose to normal tissue and decreased dose to the target. Any miss of the target, regardless of size, increases the chance for local recurrence and decreases OS. One example of an immobilization technique that has decreased patient motion is the Elekta Oncology s BodyFIX system. This device uses a fullbody, vacuum formed bag along with a plastic cover, which lies over the patient. It utilizes a

12 12 vacuum system that places a tolerable amount of pressure on the patient limiting all movement. Another commonly used immobilization device is Elekta Oncology s stereotactic body frame (SBF). The SBF also utilizes a vacuum formed bag, but this system replaces the plastic cover with a rigid frame that applies abdominal compression with adjustable pads. The SBF also utilizes an external reference system for patient alignment in both the simulation and treatment phases. 9 These types of immobilization devices combined with Image-Guided Radiation Therapy (IGRT) should provide adequate measures to decrease the possibility of a missed target. Statement of the Problem Surgical resection is the treatment of choice for stages I and II (T1-2, N0, M0) NSCLC. Individuals who are not surgical candidates due to existing comorbidities are left with one viable option, which is radiation therapy. Conventional radiation therapy for stage I and II (T1-2, N0, M0) NSCLC has a 5-year OS rate of 10%-30% with a local recurrence rate of in 55%-70% of patients. 3 Early attempts to increase OS through dose escalation were unsuccessful due to an increase in radiation-induced toxicities that accompanied these higher dose schemes. Advances in both radiation oncology and medical imaging led to the conception of SBRT. Studies have shown that increasing tumor dose while decreasing doses to surrounding structures is achievable through SBRT. However, due to the lack of long-term follow-up for more recent SBRT methods and the varying fractionation schemes used, many physicians are hesitant to utilize SBRT. An increase in the volume and standardization of long-term studies is necessary. Purpose of the Study The purpose of this quantitative study was to demonstrate the impact on medically inoperable stage I and II (T1-2, N0, M0) NSCLC patient survival when treated with SBRT compared to alternate and previously used treatment schemas such as three-dimensional (3D) and IMRT planning techniques by comparing dosimetric values. Five patients previously treated for stage I and II (T1-2, N0, M0) NSCLC were selected. Treatment plans using a coplanar 3D conformal technique of 3 and 7 field arrangements, a 9 field coplanar IMRT technique, and a field non-coplanar SBRT technique were created for each patient. Dosimetric values including the percentage of volume of the PTV receiving 90% of the prescription dose (PTV 90 ), the percentage of volume of the PTV receiving 95% of the prescription dose (PTV 95 ), the maximum dose 2 cm from the PTV in all directions (D 2cm ), the volume of an organ that is irradiated to 20 Gy or more (V 20), and Mean Lung Dose (MLD) from each plan were compared to demonstrate the benefits of SBRT.

13 13 Assumptions of the Study Patients treated using SBRT methods will have higher survival outcome statistics than those receiving standard fractionation. The comparison between plans will demonstrate that higher doses to target areas are effectively achieved through SBRT while maintaining organs at risk (OR) tolerance doses. Long and short-term radiation induced toxicity will be comparable between standard conventional radiation therapy (RT) and SBRT. Definition of Terms Biological Effective Dose (BED). Biological effective dose is the sum of the weighted dose equivalents for irradiated tissues or organs, taking into account the weighting factor of the tissue being treated, and the mean dose equivalent received by that tissue. 4 For example, one way to increase the BED is to give the same total dose normally prescribed for standard fractionation in less fractions. This is the method used by SBRT; it does not allow the cells the recovery time that is usually provided during a standard course of treatment, and is therefore more lethal. Diffusion Capacity. Diffusion Capacity is a measure of how well oxygen and carbon dioxide are transferred between the lungs and the blood. 10 D 2cm. The D 2cm is the maximum dose 2 cm from the PTV in all directions. 9 Forced Expiratory Volume. Forced expiratory volume is the maximum amount of air that can be expelled in a given number of seconds. 10 Helical Tomotherapy. Helical tomotherapy is a treatment using a linear accelerator that is constructed in the same fashion as a CT scanner. In this system, the radiation source is constantly rotating and the table is working in conjunction with the gantry. 11 Hypercapnia. Hypercapnia is a condition in which there is an abnormal increase of carbon dioxide in the arterial blood supply. 10 Hypofractionation. Hypofractionation is a technique used to increase the ability of radiation to damage a cell by decreasing the number of treatment fractions while maintaining the same or similar doses. 2 Hypoxemia. Hypoxemia is a reduction in arterial oxygenation. 10 Image Guided Radiation Therapy (IGRT). Image guided radiation therapy is the utilization of daily imaging to localize the target prior to treatment. Imaging techniques can include KeV and MV images and/or single CT scan slices that are overlaid onto images of the same type acquired during the simulation phase.

14 14 Intensity Modulated Radiation Therapy (IMRT). Intensity Modulated Radiation Therapy is a treatment that can modulate the intensity of the radiation field in order to shape the beam, increasing the conformality of dose, which reduces unwanted dose to surrounding tissues. The beam is shaped by tiny metal jaws called Multi-Leaf Collimators (MLC) that independently move creating intricate shapes for beam collimation. Intensity modulated radiation therapy treatments often use multiple coplanar fields and can have a lengthy treatment time. 4 Linear Accelerator (linac). A device that uses high-frequency electromagnetic waves to accelerate charged particles such as electrons to high energies through a linear tube. A linac can generate electron and photon beams to be used therapeutically. Typical beam energies include 6MV, 10MV, and 18 MV photons and 6MeV, 9MeV, 12MeV, 15MeV, and18mev electrons. 4 Mean Lung Dose (MLD). The mean lung dose is the mean dose received by the total volume of both lungs. 9 PTV 90 /PTV 95. The PTV 90 /PTV 95 is the percentage of volume of the PTV receiving 90%/95% of the prescription dose. 9 R 50. The R 50 is the ratio of the volume irradiated to 50% of the prescribed dose to the volume of the PTV. 9 Serial Tomotherapy. Serial tomotherapy is a treatment using a linear accelerator that is constructed in the safe fashion as a CT scanner. In this system, the radiation source performs a rotation, then the table will move and then the process repeats. 11 Stereotactic Body Radiation Therapy (SBRT). Stereotactic Body Radiation Therapy is a treatment that allows the delivery of high doses over 1-5 fractions to the tumor without increasing doses to the surrounding tissue, compared to conventional radiation therapy. This technique uses several non-coplanar, non-opposing fields with very little margin, an image guidance system used for tumor location, and strict immobilization devices. 5 Stereotactic Radiosurgery (SRS). Stereotactic radiosurgery is a radiation treatment technique that utilizes several non-coplanar arcs or beams that converge on the treatment machine s isocenter, which is centrally placed in the tumor using IGRT. This technique creates a tightly conformed dose distribution giving the ability to deliver high doses of radiation in one singular fraction. 4 Tolerance Dose (TD). An organ s tolerance dose is the maximum amount of radiation the organ can receive. 4

15 15 Tumor, Node, Metastasis (TNM) Staging. Tumor, Node, Metastasis (TNM) staging is a method of determining the stage of a cancer by utilizing the tumor size (T), nodal involvement (N), and metastasis (M). There have been several revisions to this system and it has unique subcategories for each type of treatment including lung (Tables 1-4). 12 Volume Doubling Time (VDT). Volume doubling time is the time that, if left alone, a tumor doubles in size. 13 Volumetric Modulated Arc Therapy (VMAT). Volumetric modulated arc therapy is a treatment technique using the same principles as IMRT to strictly conform dose, but it delivers the dose while the gantry is performing a rotation around the patient. 9 V 20. The V 20 is the volume of an organ that is irradiated to 20 Gy or more. 9 Limitations of the Study A limitation of the study included the limited number of patients that were treated for stage I and II (T1-2, N0, M0) NSCLC and therefore the lack of long-term follow up data available. Since surgery is the preferred treatment option for Stage I and II NSCLC, the population of patients that were irradiated was small. While this study demonstrated the ability of SBRT to decrease OR doses, the exact clinical effect of this reduction could not be calculated without further research into the relationship between doses and organ toxicity when using an increased BED. Methodology In this quantitative retrospective study 5 patients were selected and were re-planned using 3 and 7 field standard coplanar 3D conformal treatments, a 9 field coplanar IMRT, and a field non-coplanar SBRT technique using Pinnacle 3 v9.0. All data collected was charted based on dose distributions including PTV 90, PTV 95, D 2cm,V 20, and MLD. After the information was charted, descriptive statistics and percentage frequency distributions were used to demonstrate the differences in the mean, maximum, volumetric and dosimetric values between the planning techniques. Though the patient sample size was small, differences in the dose distributions of each type of treatment were sufficiently represented.

16 16 Chapter II: Literature Review The increasingly large number of patients being diagnosed with stage I and II NSCLC each year combined with the poor local control and 5 year OS rates of standard radiation therapy techniques have led to SBRT becoming an increasing trend. The ability of SBRT to decrease normal tissue toxicity while reaching doses high enough for tumor ablation, gives it the potential of being the treatment of choice in facilities that possess the technology to utilize this treatment technique. 9 The use of SBRT to treat stage I and II (T1-2, N0, M0) NSCLC has become an option due to recent technological advancements in radiation oncology. New methods of delivering radiation treatments in conjunction with improvements made to immobilization devices that reduce patient movement allow for increased precision. In some cases the additional simulation and planning time required for the precision of an SBRT treatment also means an increased waiting period to begin treatment. A study done by Murai et al 13 looked at tumor progression during this waiting period. Three hundred and nineteen patients treated between April 2004 and June 2010 were retrospectively studied. All patients were treated at 1 of 3 separate institutions in Nagoya, Japan. Computed tomography data sets on or near the date of biopsy were compared to CT data from treatment simulation. The time that elapsed between scans was considered the wait time (WT). Murai et al 13 stated that this was only an estimate due to the varying scan dates from the biopsy time frame. Tumor diameter was determined by using a lung-viewing window in the treatment planning system. The transverse CT slice containing the greatest tumor diameter was located and that diameter defined the tumor size. The growth between data sets was measured in volume doubling time (VDT). Volume doubling time is the time that, if left alone, a tumor doubles in size. For the study, the median wait time was 42 days. In 156 of the 319 patients the VDT could be calculated. These results showed that in 23% of patients with a wait time of 25 days or longer, there was no increase in tumor size. In patients with a WT greater than 4 weeks, a progression in T staging was noted. This was seen in 25 patients, 5 of which the tumor exceeded 5 cm, which was the SBRT treatment cut off point for tumor size in this study. Distant spread or lymph node involvement was not seen in any case. Once a patient has been selected to receive radiation therapy, immobilization techniques must be considered. According to Brock et al, 9 SBRT treatment times can be lengthy and it is important for the patient to not only be immobilized but to be as comfortable as possible. Creating an environment where the patient can relax their body into the immobilization device

17 17 without having to use their own strength to hold a position will reduce the patient's movement. 5 If a patient has to work to hold themselves in position they will become fatigued during the treatment and this could lead to excessive movement. The largest contributing factor to the motion that creates a problem for SBRT is respiratory-induced movement. 14 Abdominal compression is the most commonly used method to decrease patient breathing motion. Bouilhol et al 14 demonstrated that the effectiveness of abdominal suppression was dependent on tumor location. This method of practice was done by performing scans on patients with lung lesions categorized into either lower or middle/upper lesions. The patients had 2 4DCT scans both with and without abdominal suppression. Using software designed specifically to evaluate 4DCT scan data, the end-exhale and end-inhale phases were evaluated, and an ITV was created. Specific landmarks outside of the tumor motion were used to account for variables on the scan in the tumor region. Comparing these volumes demonstrated a significant reduction in motion when all of the patients were averaged together. However, when separating tumor location and comparing the lower lobe lesions with middle/upper lesions, the lower lobe lesions were significantly reduced with a mean reduction of 3.5 mm where the lesions located in the middle/upper lobes showed only a 0.8 mm reduction in movement. Some of the middle/upper lobe lesions actually showed a negative effect, where motion and ITV sizes actually increased. Bouilhol et al 14 concluded that abdominal suppression did not benefit middle/upper lobe lesions but did show beneficial reduction of the ITV in lower lobe lesions. Heinzerling et al 5 performed a similar study that utilized Elekta Oncology s SBF s suppression plate applying both median and high amounts of suppression. In this study lower lobe lung and liver lesions were used. The tumors were scanned with free breathing and the different levels of suppression using 4DCT. Using the CT scans at both peak inspiration and expiration, contours were created representing tumor and organ movement. With the increasing amount of suppression, both tumor and organ movement were reduced, with highest level of suppression tumor motion reduced to less than 1cm in all cases. Obtaining the treatment planning CT scans for SBRT is an important step prior to treatment. This step varies for each facility. Some facilities use a free breathing CT scan while others choose to use 4DCT scans to create an estimation of breathing to account for tumor motion. Currently there are treatment facilities that utilize both scan types in their treatment planning process. Han et al 15 compared plans generated on both scan types. Ten patients with stage I NSCLC were treated with SBRT. All 10 patients had peripheral lesions, which in this

18 18 study were defined as greater than 2 cm in distance from the proximal bronchial tree and trachea. An ITV was created from the 4DCT (CT AVG ) data and then fused to the free breathing CT scan (CT HEL ). Plans used for treatment purposes were generated on the CT HEL scans. Due to the ITV being created on the CT AVG scan data, the plans were just transported directly to the CT AVG scans with very few changes. Both plans for each patient were placed side by side and several dosimetric values were compared. Results showed no significant change in maximum or mean organ of risk (OR) doses. There were no clinically significant changes in location of the maximum dose point. In a study done by Purdie et al patients with early stage NSCLC received SBRT treatments utilizing IGRT. Daily cone beam computed tomography (CBCT) was used to shift the patient into position by comparing the CT data acquired during patient set-up to CT data acquired during the simulation phase. The patients were treated on an Elekta Synergy linear accelerator with an on-board kv imaging system. This system directly attaches to the treatment machine and is positioned so that the orientation of the image is orthogonal, or perpendicular, to the radiation beam. In addition to the daily CBCT, the kv imaging was used to view the tumor immediately before the beginning of the radiation treatment. Utilizing an aperture placed at the interface of the diaphragm on a 2D image, the respiratory phase was established. In one gantry rotation, approximately 640 images were acquired. These images were used to reconstruct expiration and inspiration datasets, known as a respiration correlated CBCT. These scans allow for tumor motion to be monitored on a daily basis. For 10 of the 12 patients in this study, the daily motion measured by the respiration correlated CBCT was consistent with the motion measured by the 4DCT done during simulation. However, 2 patients had discrepancies of 0.6 cm and 1 cm in the superior and inferior plane of movement. Purdie et al 16 demonstrated that tumor movement during the breathing cycle on the day of treatment may consistently differ from the movement captured during planning. This study demonstrated that using tumor motion acquired during a 4DCT scan may not be adequate to gauge tumor motion for an entire course of RT, SBRT or standard fractionation. Respiration correlated CBCT combined with adaptive radiation plans could be the next step in the treatment of NSCLC with SBRT. Brock et al 9 conducted a study to evaluate different types of treatments on patients who had been previously irradiated. Planning CT scans from 5 patients who had a tumor diameter of less than 3 cm and T1N0 were selected. Brock et al 9 used an in-house planning program that utilized a complex algorithm that generated coplanar and non-coplanar inverse plans of 3, 5, 7

19 19 and 9 fields. All plans utilized 6MV and a 6 mm margin around the PTV to account for penumbra. The PTV in every case was the GTV plus a 1 cm margin. The prescription used for all plans was 8 fractions of 7.5 Gy for a total dose of 60 Gy prescribed to central point in the PTV. Two volumetric modulated arc therapy (VMAT) plans were generated using the same algorithm, one with the standard 6 mm margin on the PTV that was used previously and one with a 2 mm margin. In order to monitor lung toxicity, Brock et al 9 used the fraction of lung tissue treated to 20 Gy or higher (V 20 ) and the MLD. The BED of SBRT was increased from the standard fractionation and could not be directly compared. V 20 is typically limited to 35% or less. A worst case scenario demonstrated that the V 20 delivered over the conventional 32 fractions would scale to a V 11 when using 8 fractions. Tolerance doses for other organs were converted using the same scaling. Other doses that were monitored included R 50 and D 2cm. Brock et al 9 also considered treatment times in this study as many patients who are not eligible for surgery, due to poor health, have difficulty remaining still for the length of treatment. All of the planned treatments were performed on a phantom and the data was compared. The results showed that utilizing VMAT and 5-7 non-coplanar beams improved dose coverage to the tumor while lowering the amount of lung treated with the higher dose when compared to the coplanar plans. As expected, the VMAT treatments were drastically faster than the others and the average time for a VMAT plan was 2 minutes. The total time the patient would be in the room for the VMAT treatment was about 22 minutes compared to minutes with the other planning techniques. Volumetric modulated arc therapy was declared the optimal treatment technique over the noncoplanar plans due to its reduction of time the patient was on the treatment table. There are currently several different treatment modalities being used in the delivery of SBRT including coplanar and non-coplanar IMRT, RapidArc VMAT, and tomotherapy. The use of tomotherapy was studied by Fuss et al 11. The study compared helical and serial tomotherapy delivery methods for the treatment of SBRT. Twelve patients were treated with SBRT on a helical tomotherapy unit from December 2005 until the time of the study in Helical tomotherapy makes use of CT type of delivery where they radiation source is constantly rotating and the table is working in conjunction with the gantry. 11 These patients were also planned using the serial tomotherapy method, which is a system where the radiation source performs a rotation, then the table will move and the process will repeat. 11 Only 6 of the 12 patients had lung lesions, and in 1 of the 6 cases the tumor was metastatic. Patients were immobilized using the BodyFix System. Patients received treatment planning computed tomography (TPCT) scans of free

20 20 breathing and a 4DCT. The 4DCT data was used to create a maximum intensity projection (MIP). The free breathing TPCT and 4D MIP were both utilized for target and OR delineation. In order to create the PTV, a 5 mm margin was added to the 4D GTV. Both plans were generated from these data sets. The fractionation scheme for the patients with inoperable NSCLC was 12 Gy per fraction for 3 fractions. The plans were compared using total monitor unit (MU), pencil beam dimensions used, volumes of the GTV and PTV as well as average and maximum doses to both structures. Delivery times were also monitored and compared. The average MU used for serial tomotherapy was 10,108. That number rose to 17,170 for the helical treatments. The mean treatment time between the 2 types differed by 7 minutes with helical being the slower of the two. The average times for the serial and helical tomotherapy treatments were 41.5 minutes and 48.3 minutes respectively. The doses for the helical plans were significantly higher than the doses of the serial treatment plans. The maximum dose escalated from 135% to 187% for the helical plans. Mean doses to the GTV and PTV went from 119% and 112% with the serial plans to 159% and 129% with the helical plans. This increase in radiation to the patient suggests that the use of serial tomotherapy for SBRT could result in a better treatment outcome than helical tomotherapy Due to the increased amounts of radiation delivered by SBRT treatments, the analysis of radiation-induced toxicities (Table 5) is valuable. Any correlations that are created between dosimetric planning parameters and toxicity may assist in a reduction in frequency of complications. A study by Paludon et al 17 demonstrated changes in dyspnea were related to specific dosimetric values such as MLD and V 20. Thirty-two patients were initially enrolled in a phase II SBRT study. This study delivered a dose of 45 Gy in 3 fractions over 5-8 days. Four patients were excluded due to a nonrelated decline in medical condition, leaving 28 to be studied. Mean lung dose was calculated for the ipsilateral lung and total lung. Both of these volumes excluded the CTV. Paludan et al 17 found the MLD for the ipsilateral lung and total lung to be 5.4 Gy and 3.6 Gy respectively. An increase in dyspnea was recorded in 11 out of the 28 patients. However, no direct correlation was found between MLD and the increase in dyspnea. Timmerman et al 3 also performed a study dealing with toxicity in SBRT outcomes. Positive signs of increased local control and OS were shown using SBRT but due to the general health and age of most patients who could not undergo surgery, many died from health issues not related to the cancer. Seventy patients who received SBRT for stage I NSCLC with tumors no larger than 7 cm were studied. All patients were immobilized using a SBF. The GTV was

21 21 identified on the TPCT and expanded 5 mm transversely and 10 mm longitudinally to create the PTV. Patients were given 60 to 66 Gy over 3 fractions over a period of 1-2 weeks. Timmerman et al 3 found that grade 1-2 toxicity were present in 58 of the 70 cases. Common toxicities listed were radiation pneumonitis, fatigue, and musculoskeletal discomfort. Eight patients developed grade 3-4 toxicity including pleural effusion, decline in pulmonary function, skin reactions, and pneumonias. Six patients died from complications believed to be associated with the SBRT treatment. Four died from bacterial pneumonia, 1 patient from pericardial effusion, and one from a locally recurring tumor. A study done by Bauman et al 18 analyzed the outcomes of a prospective phase II trial using SBRT for stage I NSCLC in both patients who were medically inoperable or who refused surgery. Sixty patients were recruited at 7 different treatment centers throughout Sweden, Norway and Denmark between 2003 and Tumor histology was required in order for patients to be part of the study. High-risk patients were required to have a biopsy as well as consecutive CT scans 3 months apart. These scans needed to show tumor progression or other evidence of malignancy. Positron emission tomography scans became standard during the years covered in this trial and were used for the diagnosis of a majority of patients without a biopsy. Patients with centrally located tumors that were adjacent to the esophagus, trachea, or main bronchus were excluded from the trial due to doses surpassing organ TDs. Dosimetric margins included an expansion of 1 to 2 mm of the GTV to create the CTV. The CTV was then expanded 5 to 10 mm transversally and 10 mm longitudinally to create the PTV. Patients were immobilized using a SBF. While the patient was immobilized, they underwent fluoroscopy to record respiratory motion. Twenty-seven patients were outside of the motion tolerance and required abdominal compression. Patients received a dose of 45 Gy over 3 fractions given every other day using 6 MV. Median treatment length was 5 days. Post SBRT, patients were evaluated at 6 weeks and then in 3 month intervals for the first 18 months. The patients were also brought back at 24 and 36 months for follow-up. A chest CT scan or x-ray was performed at every follow-up. Cases of suspected relapse required a PET scan to be performed. Bauman et al 18 found that there was a progression free survival rate of 52% at 36 months. Median survival was 40.6 months for the total group. Twenty-seven patients died during the observation period, of which 7 were a result of the lung cancer. No significant difference was found in the outcome when comparing T1 and T2 lesions, although the tumors that were less than 2 cm in diameter had a very favorable outcome. The estimated local control rate was 92% at 3 years. Eight percent of patients who

22 22 experienced a local relapse had a mean GTV volume that was larger than those who did not experience local relapse. Throughout observation, toxicity was mild. Seventeen patients experienced grade 3 and 4 toxicity and 8 of those were still alive at the conclusion of the study. Twenty-five percent of patients did not experience any pulmonary side effects. Another study of treatment outcomes performed by Grills et al 19 analyzed the outcomes of SBRT and wedge resection for Stage I NSCLC. One hundred twenty-seven patients were treated; 69 were treated with a wedge resection and 58 using SBRT. The 69 patients who underwent wedge resection were not eligible for a lobotomy according to their thoracic surgeon. An experienced thoracic surgeon determined the extent of wedge resection. The intent of wedge resection was curative with negative margins, but limited by each patient s tolerance to the procedure. These patients were treated from February 2003 through February 2008 at William Beaumont Hospital. The 58 patients treated with SBRT were part of a prospective phase II trial that began in November Five percent of the SBRT patients were eligible for surgery but refused. Patients were immobilized using a SBF, Body Fix or alpha cradle. Under fluoroscopy, tumor motion was monitored and, in patients with excessive motion or poor visualization, a 4DCT was used. The CT data was fused with a PET scan in order to create a plan. A limited number of couch angles were used along with 6-9 beam angles to generate the SBRT plans. An ITV was created and given a 4 mm expansion in order to create the CTV. A 5 mm expansion in all 3 dimensions was added to the CTV to create the PTV. The treatment prescription was 12 Gy per fraction for either 4-5 fractions. For observation purposes, patients were given a chest x-ray (CXR) with or without a CT scan. Post SBRT, patients were given a PET-CT at 6 weeks, 16 weeks, and 12 months. In cases that the PET-CT suggested reoccurrence or progression of the cancer, biopsies were obtained. Grills et al 19 found that in most categories there was very little difference between wedge resection and SBRT. Stereotactic body radiation therapy did however show an increase in local control. Wedge resection was superior in both regional recurrence and distant metastasis, but this is believed to be a result of the increased existing comorbidities on the SBRT patients. When looking at cancer deaths specifically, OS was nearly identical. Complications from SBRT included a grade 2 or worse radiation induced pneumonitis in 11% of patients. Eleven percent of patients developed rib fractures and in 38%, a grade 1 skin toxicity was developed. For wedge resection, a median time of 5 days was spent in the hospital post surgery with a 30 day or less readmission rate of 10%. In 3% of patients, a chest tube was required.

23 23 Joyner et al 20 performed a study of treatment outcomes on centrally located lung tumors. In this retrospective study, all 9 patients who were treated for centrally located lung lesions at the University of Texas Health Science Center since August 2001 were selected. Stereotactic body radiation therapy was used in all 9 cases, which included both metastasis and recurrent NSCLC. Joyner et al 20 defined a centrally located tumor as all or part of the tumor being less than 2 cm from the following mediastinal structures: right and left lower and upper lobe bronchi, right and left main bronchi, carina, intermedius and lingular bronchus. Patients were prescribed 12 Gy per fraction for 3 fractions to at least 95% of the PTV. The PTV was created by expanding the GTV by 5 mm transversely and 10 mm longitudinally. Patients were simulated and treated using the Body Fix system. Patients were brought in at 6 weeks post SBRT for a follow up and then in 3 month intervals for 1 year and then every 4-6 months beyond that. In most cases patients received CT or PET-CT scans at the time of the follow-up. Joyner et al 20 examined OS rate, tumor locations and volumes, and patient follow-up data. They found that hilar lesions were generally larger than parenchymal lesions. In all 9 cases, local tumor control was confirmed and 3 were still living at the conclusion of the study. According to the imaging studies, complete or partial tumor response occurred in all cases. Only 1 patient showed significant toxicity in the form of a narrowing at the right lower lobe bronchus however, no further treatment was required for this patient. Palma et al 21 performed a study observing the impact of SBRT on patients age 75 and up over the course of 3 different time frames. The time frame assigned was based on diagnosis and was divided as such: period A ( ), period B ( ), and period C ( ). During period A, SBRT was not used in the region. Period B began the introduction of SBRT that gradually increased through the time period. Period C was a period in which the local clinics had full access to SBRT. Although the exact number of cases for each year was not available, these periods were selected to represent specific percentages of SBRT use. During period A there was no SBRT use, during period B very few facilities used SBRT as it was in developmental phases, and during period C most of the facilities in the area were using SBRT. All information for this study was taken from the Amsterdam Cancer Registry, which is a population-based cancer registry. This registry did not contain specific details on RT such as dose or fractionation. Radiotherapy treatments were also not divided into categories of standard RT and SBRT. The median age for all time periods at time of diagnosis for patients undergoing RT was 80 years old. Estimates showed an increase in median survival by 5 months from period A to period C. During

24 24 period A, median survival for surgery was 34 months compared to 20 months of RT. By period C, surgery had increased 2 months to 36, but RT showed significant improvement of 6 months to 26 months. Patients who decided to not undergo either treatment modality had a poor median survival of 6 months with 21% dying within 30 days of diagnosis. Bauman et al 22 performed a study of the treatment results of 138 SBRT cases attempting to demonstrate the key factors for the delivery of SBRT at a highly efficient level. Beginning in 1996, 141 Stage I NSCLC patients were selected to receive SBRT at 4 separate institutions in Sweden and 1 institution in Denmark. The majority of patients were deemed medically inoperable due to various comorbidities, however 4% refused surgery and chose to undergo SBRT. The patients received TPCT scans and patients who were not at risk of a fatal pneumothorax received biopsies to establish histology. Patients were treated with the use of the SBF and an energy of 6MV. While the patient was immobilized in the SBF, fluoroscopy was used to monitor tumor motion. If the motion exceeded 5 mm, upper abdominal compression was applied in order to control motion. Bauman et al 22 used previous experience and the reproducibility of the SBF to determine dosimetric margins. A 5 mm margin in the transverse plane and 10 mm in the longitudinal plane were added to the CTV creating the PTV. The fractionation schemes used were 2-4 fractions with doses of 10 Gy to 20 Gy per fraction. Fractions were given 2-3 days apart. Using 5-9 beams 28 Gy to 30 Gy was delivered to the periphery of the PTV. This allowed central parts of the PTV to receive up to 150% dose. Bauman et al 22 separated the post SBRT evaluation into 4 categories. A complete response (CR) was no visible tumor. If the tumor reduced in size by 50% or more it was considered a partial response (PR). The stable disease (SD) category were tumors that either had a >50% decrease in size or a <25% increase in size. Lastly, any tumor with 25% increase in size was considered local failure. Based on CT scans performed post SBRT with a median period of 16.3 months, the CR and PR groups totaled 61%. Stable disease was noted in 36% of the patients. Local control failed in 12% of the patients over months post SBRT. Local control failed more often among T2 lesions and was more often found in centrally located tumors. Statistically there was an advantage in survival at the higher dose schemes. Sixty-five percent of the patients died during follow up. Of that group, 60% did not die from lung cancer. Three year OS was 52% and 5 year OS was 26%. Overall toxicity was mild and 60% of patients did not complain of side effects. With SBRT treatment outcomes becoming more favorable, the obvious question becomes how it compares to surgery in individuals who do not have the pre-existing co morbidities that

25 25 prevent surgery. Onishi et al 23 performed a study in attempt to answer that question. In this retrospective study, 87 patients with T1-T2 N0M0 NSCLC were selected from 14 major Japanese institutions. These 87 patients were medically operable and given surgery as an option but selected SBRT as their treatment option. Due to the large number of facilities used, SBRT treatment techniques were not the same for all facilities. Five requirements were created for treatments in order for them to be considered for the study. These 5 requirements were suppression of respiratory motion to less than 5 mm, 3 mm or less slice thickness for TPCT scans, a dose per fraction of at least 5 Gy, a set-up error of 5 mm or less using IGRT, and a treatment plan using multiple non-coplanar static ports or dynamic arcs. No restrictions were placed on tumor location. Typical treatment schemes were 75 Gy in 10 fractions and 48 Gy in 4 fractions. The median BED was calculated at 116 Gy. Evaluation focused on survival rate, local control and toxicity. In most cases, follow up post SBRT was 4 weeks and then every 1-3 months. For this study Onishi et al 23 defined a CR as the tumor completely disappearing or being replaced by fibrotic tissue. A 30% or greater decrease of the maximum diameter on the crosssection scan was classified a PR. Tumor growth continuing more than 6 months post SBRT was considered local recurrence for this study. A CR was documented in 32% of patients while a PR was noted in 49%. Five year OS for all cases was 69.5%. The 5-year case specific survival rate was 76.1%. Toxicities up to grade 3 were noted with the majority (70%) being grade 1. The grade 3 toxicities totaled 9.2%, and rib fractures were noted in 4.6% of cases.

26 26 Chapter III: Methodology The increasingly large number of patients being diagnosed with stage I and II NSCLC each year combined with the poor local control and 5 year OS rates of standard radiation therapy techniques have led to SBRT becoming an increasing trend. The ability of SBRT to decrease normal tissue toxicity while reaching doses high enough for tumor ablation, gives it the potential of being the treatment of choice in facilities that possess the technology to utilize this treatment technique. 9 The use of SBRT to treat stage I and II (T1-2, N0, M0) NSCLC has become an option due to recent technological advancements in radiation oncology. New methods of delivering radiation treatments in conjunction with improvements made to immobilization devices that reduce patient movement allow for increased precision. Subject Selection and Description Five patients with Stage I or II NSCLC with a tumor size of 7 cm or less were selected. Tumor location included peripheral lesions only. Some studies suggest that centrally located tumors, less than 2 cm from the following mediastinal structures: right and left lower and upper lobe bronchi, right and left main bronchi, carina, intermedius and lingular bronchus, should not be treated due to increased complications from radiation-induced toxicities. 18,20 The dose scheme used for both 3D conformal treatments and IMRT was 70 Gy delivered daily over 35 fractions. Patients planned with SBRT were given 10 Gy per fraction for 5 fractions with minimum interval of 48 hours. Instrumentation The patients were scanned on a large bore Toshiba Aquilion CT scanner with 3 mm slices. Patients were positioned supine using a Vac-Lok bag with their arms raised above their heads. The Vac-Lok bag extended from the top of the patients shoulders to mid-thigh and a breathing suppression band was used. The plans were completed with the Philips Pinnacle 3 v9.0 treatment planning system. Data Collection Procedures All 4 treatment plans were generated from the same scan for each patient. The physician contoured the full inspiration, full expiration and free breathing GTVs. Those volumes were then fused and combined using MIM vista. The normal breathing GTV with a 2 cm margin was used to create the PTV for the 3 field 3D conformal plan. The ITV with a 1.5 cm margin was used to create the PTV for the 7 field 3D conformal plan. A 0.7 cm block margin was used for both 3D conformal techniques. The IMRT plans utilized the same PTV as the 7 field 3D conformal plans.

27 27 A 1 cm margin on the ITV created the PTV used for SBRT planning. No block margin was used for the SBRT planning. The medical dosimetrist contoured all normal structures. All plans were created by the same medical dosimetrist to ensure consistency in the plans. Upon completion of all plans contour volumes were created including the 90% and 95% isodose lines which were used in the creation of the PTV 90 and PTV 95. A contour representing 50% of dose was created for each plan in order to calculate the R 50. The V 20 was created by using the contour expansion tool in the treatment planning software to subtract the lung contour from the contour created by the 20 Gy isodose line. Plan summaries were printed for each plan one all contours were created. The data was collected from the plan reports and charted. Data analysis After all of the treatment plans were completed, the data was recorded based on dose distribution including PTV 90, PTV 95, D 2cm,V 20, and MLD. For each patient, an Excel spreadsheet was created with the numerical data of each dosimetric value for all plans. The categorical values for all 5 patients were recorded and a mean value was entered for each category created. Descriptive statistics and percentage frequency distributions were used to demonstrate the differences in the mean, maximum, volumetric, and dosimetric values between the planning techniques. Due to the differing fractionation schedule s effect on organ tissue it was difficult to directly compare doses to a suspected or desired outcome. Limitations A limitation of the study included the limited number of patients that were treated for stage I and II (T1-2, N0, M0) NSCLC and therefore the lack of long-term follow up data available. Since surgery is the preferred treatment option for Stage I and II NSCLC, the population of patients that were irradiated was small. While this study demonstrated the ability of SBRT to decrease OR doses, the exact clinical effect of this reduction could not be calculated without further research the relationship between doses and organ toxicity when using an increased BED. Summary Advancements made in the delivery of radiation treatments in conjunction with improvements made to immobilization devices that reduce patient movement allow for increased precision, which allows for an increase in prescription doses. Five patients with Stage I or II NSCLC with a tumor size of 7 cm or less were selected. Tumor location included peripheral lesions only. Four plans were created for each patient. The techniques used were 3 and 7 field 3D

28 28 conformal coplanar, 9field coplanar IMRT, and a field non-coplanar SBRT technique. All 4 treatment plans were generated from the same scan for each patient. Plan summaries were printed for each plan, once all contours were created. The data was collected from the plan reports and charted. The data was compared and the results were demonstrated with descriptive statistics showing the differences in dosimetric values such as V 20, PTV 95, critical structure doses, spinal cord dose, heart dose, and D 2cm. Continuing research demonstrating the impact of SBRT treatments on patient outcomes is necessary to improve and standardize SBRT. Increasing the amount of long term follow-up information and treatment plan comparisons can provide information needed to select the most effective dose fractionation scheme.

29 29 Chapter IV: Results The ability of SBRT to decrease normal tissue toxicity while reaching doses high enough for tumor ablation gives it the potential of being the treatment of choice in facilities that possess the technology to utilize this treatment technique. 9 The use of SBRT to treat stage I and II () NSCLC has become an option due to recent technological advancements in radiation oncology. This study demonstrated the ability of SBRT to decrease doses to normal tissues while escalating dose to the tumor. Item Analysis R 50 After the completion of all treatment plans, the data was compared and trends were identified. Figure 1 shows the R 50 for all 4 plans for each patient. The data demonstrates a decrease in the R 50 as the planning technique increases in difficulty from the 3 and 7 field 3D conformal technique to the IMRT plans. A decrease in R 50 represents a reduction in doses to surrounding normal tissues. The data shows an average reduction of 10.4% in R 50 when comparing 3 fields to 7 fields and a reduction of 28.1% when comparing the 7 field plan to IMRT. An increase in R 50 is shown when comparing the SBRT plans to the IMRT plan. The average increase was 21.2% with a maximum increase of 42.6% and a minimum increase of 6.7%. D 2cm Figure 2 shows a significant reduction in D 2cm as the treatment plans progress from simple to complex. A decrease in D 2cm demonstrates the ability of a treatment technique to tightly conform the prescription dose to the treatment target. Utilizing a 7 field 3D plan showed an average dose reduction of 15.1% with a maximum D 2cm reduction of 22.7% when compared to the 3 field 3D plan. The IMRT plan showed an average decrease of 3.7%, with a maximum reduction of 20.2% and an increase of 32.4% in one case when compared to the 7 field 3D plan. The field non-coplanar SBRT plans demonstrated an average reduction of D 2cm of 34.1% with the maximum reduction being 51.9% and a minimum reduction of 22.9% when compared to the 9 field IMRT plan. PTV 90 /PTV 95 PTV 90 and PTV 95 coverage represents how well a RT plan has covered the treatment target with the prescription dose. An increase in the PTV 90 and PTV 95 indicates an increase in the dose coverage of a plan. Figure 3 shows the PTV 90 for all plans. The data demonstrates a

30 30 small increase of 6.9% in PTV 90 when progressing from the 7 field 3D treatments to the IMRT treatments. Comparing the SBRT treatments to IMRT demonstrated a 0.9% increase in PTV 90. Comparing the data from the 7 field 3D conformal plan to the 3 field 3D conformal plan shows an average reduction in PTV 90 of 3.8% with a maximum reduction of 12.9% and a maximum increase of 3.1%. Changes in PTV 95 demonstrate a similar trend to the PTV 90 (Figure 4). An average increase of 14.4% was seen when comparing IMRT to the 7 field 3D plan. Comparing the SBRT plans to IMRT showed an average increase of 4.8%. Similar to PTV 90, PTV 95 shows a reduction in PTV 95 when comparing the 7 field 3D plans to the 3 field 3D plans with the average decrease being 6.1% MLD In Figure 3 a decreasing trend of MLD is shown as treatment techniques progress from simple to more complex with a large change occurring when comparing IMRT to SBRT. A decrease in MLD of 42.3% with a maximum reduction of 57.3% and minimum of 34.6% is demonstrated. V 20 Figure 6 shows the lung V 20 for all patients. When comparing the results of the 7 field 3D plan to the 3 field 3D plan, an increase in size of the V 20 was shown. The average change was an increase in volume of 8.4%; the largest volume increase was 25.7% while one volume showed a decrease of 9.3%. An average reduction in volume of 16.9% was shown in the IMRT plan when compared to the 7 field 3D plan. The largest reduction in V 20 for the IMRT plan was 35.3% with the minimum change being a reduction of 2.2% when compared to the 7 field 3D plan. The SBRT plans demonstrated a significant reduction in the V 20 of the lung when compared to the IMRT plans, with the average reduction being 49.7%. The SBRT plans resulted in a maximum decrease of 72.2% and a minimum decrease of 41.2%. Spinal cord dose Figure 7 shows the mean spinal cord dose for all plans. The data shows a trend of the mean cord dose decreasing when comparing the SBRT plans to the other plans. The SBRT plans showed an average reduction of 48.3% in mean spinal cord dose when compared to the data from the IMRT plans with the maximum reduction being 94.1%. Figure 8 demonstrates a similar trend with the maximum cord dose. The average decrease in maximum cord dose was 45.4% when comparing SBRT plans to IMRT. Heart dose

31 31 Similar to the cord doses, the mean and maximum heart doses show a significant decrease when comparing the SBRT to the standard fractionation planning (Figure 9). The data shows a reduction in mean heart dose of 40.9% when comparing SBRT to IMRT. Stereotactic body radiation showed a maximum reduction in mean heart dose of 63.6% and a minimum reduction of 20.1%. The maximum heart dose results also demonstrate a trend in dose reduction when comparing SBRT to IMRT (Figure 10). Summary The results obtained from this study demonstrated the ability of SBRT to increase the percentage of the PTV receiving 90%-95% of the prescription dose. The field noncoplanar used for SBRT reduced V 20, MLD, cord dose, heart dose, and D 2cm. The reduction in these values could potentially demonstrate a reduction in normal tissue toxicity when using SBRT to treat stage I and II (T1-2, N0, M0) NSCLC. Additional studies directly relating the percentage reduction of these values to specific treatment toxicities are suggested.

32 32 Chapter V: Discussion The increasingly large number of patients being diagnosed with stage I and II NSCLC each year combined with the poor local control and 5 year OS rates of standard radiation therapy techniques have led to SBRT becoming an increasing trend. The ability of SBRT to decrease normal tissue toxicity while reaching doses high enough for tumor ablation, gives it the potential of being the treatment of choice in facilities that possess the technology to utilize this treatment technique. The use of SBRT to treat stage I and II (T1-2, N0, M0) NSCLC has become an option due to recent technological advancements in radiation oncology. New methods of delivering radiation treatments in conjunction with improvements made to immobilization devices that reduce patient movement allow for increased precision. There have been many changes in the recent years in the treatment of NSCLC with RT. The use of an ITV derived from combining GTVs from scans at different phases in the respiratory cycle allows the tumor to be accurately delineated throughout the breathing cycle. Improvements in patient immobilization and the use of IGRT have allowed a reduction in the PTV, while not increasing the potential for missing the target. This reduction in PTV size also decreases the doses to uninvolved tissue. These advancements have allowed an increase in dose and decrease in the amount of fractions required for RT borrowing from the success of many SRS programs worldwide. Stereotactic body radiation therapy has been used primarily to treat patients with stage I and II NSCLC whom do not meet requirements for surgery. Many comorbidities, including forced expiratory volume, reduced diffusion capacity, hypoxemia and/or hypercapnia, can exclude a patient from being eligible for surgery. Because the individuals do not qualify for surgical resection they are given the option for RT. The results of the current study, along with the aforementioned studies in Chapter 2 support the ability of SBRT to increase dose to the tumor while decreasing doses to normal tissues. For example, the increase in PTV 90 and PTV 95 shown in the current study correlates back to the PTV 90 and PTV 95 increase shown by Brock et al 9 whose research demonstrated that increasing the number of fields and utilizing a non-coplanar technique would increase the percentage volume of the tumor that receives higher doses. This increase can be tied directly to increases in patients OS demonstrated by Bauman et al. 22 Bauman et al 22 linked the increased tumor doses achievable with SBRT to an increase in outcomes such as OS and a reduction in local recurrence. Bauman et al 22 suggest these improved treatment outcomes could demonstrate the potential for SBRT to rival surgical outcomes in operable patients.

33 33 Papiez et al 2 discussed the escalation in lung toxicity during prior attempts at increasing doses using standard fractionations. The inability for patients to tolerate the side effects caused by the increased normal tissue toxicity halted these earlier attempts. However, contrary to the study by Papiez et al 2, the current study demonstrated a reduction in MLD, V 20, D 2cm, and R 50 which supports the fact that SBRT has the ability to overcome previous complications in attempts to hypofractionate lung treatments for stage I and II (T1-2, N0, M0) NSCLC discussed by Papiez et al. 2 Limitations A limitation of the study included the limited number of patients that were treated for stage I and II (T1-2, N0, M0) NSCLC and therefore the lack of long-term follow up data available. Since surgery is the preferred treatment option for Stage I and II NSCLC, the population of patients that were irradiated was small. While this study demonstrated the ability of SBRT to decrease OR doses, the exact clinical effect of this reduction could not be calculated without further research the relationship between doses and organ toxicity when using an increased BED. Conclusions Surgical resection is the treatment of choice for stages I and II (T1-2, N0, M0) NSCLC. Individuals who are not surgical candidates due to existing comorbidities are left with one viable option, which is RT. Conventional RT for stage I and II (T1-2, N0, M0) NSCLC has a 5-year OS rate of 10%-30% and local recurrence experienced in 55%-70% of patients. 3 Early attempts to increase OS through dose escalation were unsuccessful due to an increase in radiation-induced toxicities that accompanied the higher dose schemes. Advances in both radiation oncology and medical imaging led to the conception of SBRT. Studies have shown that increasing tumor dose while decreasing doses to surrounding structures is achievable through SBRT. The results of this study demonstrated that treating inoperable Stage I or II NSCLC with SBRT increased dose to the tumor, decreased doses to OR and improved overall dose conformity. The average decrease in D 2cm for all patients when comparing a 3 field coplanar 3D plan to SBRT demonstrated the ability of SBRT to tightly conform prescription dose to the PTV. While this study demonstrated the ability of SBRT to decrease OR doses, the exact clinical effect of this reduction could not be calculated without further research. The varying levels of BED due to differences in fractional and total doses prevented directly relating the average increase or decrease in V 20 to an equal change in compromised lung tissue. Although the direct link from

34 34 dosimetric values could not be made for OR, the dosimetric increase in tumor dose combined with the increase in BED could be directly linked to tumor cell death. The average increase in PTV 90 and PTV 95 when comparing SBRT to 3D conformal planning demonstrated the ability of SBRT to increase dose to the PTV. Recommendations A similar study should be conducted with an increase in sample size. This could be done by increasing the length of the study dates, using a larger academic institution to gather patients, or increasing the study to include multiple clinics. Additional research documenting patient outcomes through the collection of clinical follow up details should be performed to demonstrate the clinical relevance of these findings. The results of this study could be used to determine clinical outcomes through more extensive research comparing the dosimetric values established in this study along with previous studies linking radiation doses to expected cellular effects. A study that specifically relates dose equivalents of both standard and hypofractionated RT to specific toxicities and treatment outcomes would be beneficial.

35 35 Tables Table 1. Tumor (T) descriptors of the TNM staging system 12 T: Tumor Description TX Primary tumor cannot be assessed, or tumor proven by the presence of malignant cells in sputum or bronchial washings but not visualized by imaging or bronchoscopy T0 No evidence of primary tumor Tis T1 T1a T1b T2 T2a T2b T3 T4 Carcinoma in situ Tumor < 3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus (i.e., not in the main bronchus) Tumor < 2 cm in greatest dimension Tumor > 2 cm but < 3 cm in greatest dimension Tumor > 3 cm but < 7 cm or tumor with any of the following features (T2 tumors with these features are classified T2a if < 5 cm): Involves main bronchus, > 2 cm distal to the carina Invades visceral pleura Associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung Tumor > 3 cm but < 5 cm in greatest dimension Tumor > 5 cm but < 7 cm in greatest dimension Tumor > 7 cm or one that directly invades any of the following: Chest wall (including superior sulcus tumors), diaphragm, phrenic nerve, mediastinal pleura, parietal pericardium Tumor in the main bronchus < 2 cm distal to the carina but without involvement of the carina Associated atelectasis or obstructive pneumonitis of the entire lung Separate tumor nodule(s) in the same lobe Tumor of any size that invades any of the following: Mediastinum, heart, great vessels, trachea, recurrent laryngeal nerve, esophagus, vertebral body, carina Separate tumor nodule(s) in a different ipsilateral lobe

36 36 Table 2. Node (N) descriptors of the TNM staging system 12 N: Nodes Description NX Regional lymph nodes cannot be assessed N0 N1 N2 N3 No regional lymph node metastasis Metastasis in ipsilateral peribronchial and/or ipsilateral hilar lymph nodes and intrapulmonary nodes, including involvement by direct extension Metastasis in ipsilateral mediastinal and/or subcarinal lymph node(s) Metastasis in contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s) Table 3. Metastasis (M) descriptors of the TNM staging system 12 M: Metastases Description MX Distant metastasis cannot be assessed M0 M1 M1a M1b No distant metastasis Distant metastasis Separate tumor nodule(s) in a contralateral lobe tumor with pleural nodules or malignant pleural/ pericardial effusion Distant metastasis

37 37 Table 4. Lung tumor staging 12 T and M Descriptors N0 Staging N1 Staging N2 Staging N3 Staging T1a I A II A III A III B T1b I A II A III A III B T2a I B II A III A III B T2b II A II B III A III B T3 II B III A III A III B T4 III A III A III B III B M1a IV IV IV IV M1b IV IV IV IV Table 5. Lung Toxicity Grades as defined by the RTOG Grade Definition 1 Mild dry cough or DOE not requiring clinical intervention 2 Cough requiring narcotic antitussives or dyspnea not at rest 3 Severe cough not responsive to narcotics or dyspnea at rest; intermittent oxygen or steroids may be required 4 Continuous oxygen or assisted ventilation 5 Fatal

38 38 Figure 1. R 50 shown for all plans in ratio Figures

39 Figure 2. D 2cm shown for all plans in dose (Gy) 39

40 Figure 3. PTV 90 for all plans shown in percentage of volume 40

41 Figure 4. PTV 95 for all plans shown in percentage of volume 41

42 Figure 5. MLD shown for all plans in dose (Gy) 42

43 Figure 6. Lung V 20 shown for all plans in volume (cc) 43

44 Figure 7. Mean spinal cord dose shown for all plans in dose (Gy) 44

45 Figure 8. Maximum spinal cord dose shown for all plans in dose (Gy) 45

46 Figure 9. Mean heart dose shown for all plans in dose (Gy) 46

47 Figure 10. Maximum heart dose shown for all plans in dose (Gy) 47