- Email: [email protected]
Computed Tomography Image-Guided Intensity-Modulated Radiation Therapy for Cervical Carcinoma With Positive Para-Aortic Lymph Nodes
Int. J. Radiation Oncology Biol. Phys., Vol. 72, No. 4, pp. 1134–1139, 2008 Copyright Ó 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/08/$–see front matter
doi:10.1016/j.ijrobp.2008.02.063
CLINICAL INVESTIGATION
Cervix
PROSPECTIVE CLINICAL TRIAL OF POSITRON EMISSION TOMOGRAPHY/ COMPUTED TOMOGRAPHY IMAGE-GUIDED INTENSITY-MODULATED RADIATION THERAPY FOR CERVICAL CARCINOMA WITH POSITIVE PARA-AORTIC LYMPH NODES JACQUELINE ESTHAPPAN, PH.D.,* SUMMER CHAUDHARI, M.S.,* LAKSHMI SANTANAM, PH.D,* SASA MUTIC, M.S.,* JEFFREY OLSEN, B.S.,* DUSTEN M. MACDONALD, M.D.,* DANIEL A. LOW, PH.D.,* ANURAG K. SINGH, M.D.,y AND PERRY W. GRIGSBY, M.D.* * Department of Radiation Oncology, Washington University School of Medicine, St. Louis, MO; and y Radiation Oncology Branch, National Cancer Institute, Bethesda, MD Purpose: To describe a more aggressive treatment technique allowing dose escalation to positive para-aortic lymph nodes (PALN) in patients with cervical cancer, by means of positron emission tomography (PET)/computed tomography (CT)–guided intensity-modulated radiation therapy (IMRT). Here, we describe methods for simulation and planning of these treatments and provide objectives for target coverage as well as normal tissue sparing to guide treatment plan evaluation. Methods and Materials: Patients underwent simulation on a PET/CT scanner. Treatment plans were generated to deliver 60.0 Gy to the PET-positive PALN and 50.0 Gy to the PALN and pelvic lymph node beds. Treatment plans were optimized to deliver at least 95% of the prescribed doses to at least 95% of each target volume. Dose–volume histograms were calculated for normal structures. Results: The plans of 10 patients were reviewed. Target coverage goals were satisfied in all plans. Analysis of dose– volume histograms indicated that treatment plans involved irradiation of approximately 50% of the bowel volume to at least 25.0 Gy, with less than 10% receiving at least 50.0 Gy and less than 1% receiving at least 60.0. With regard to kidney sparing, approximately 50% of the kidney volume received at least 16.0 Gy, less than 5% received at least 50.0 Gy, and less than 1% received at least 60.0 Gy. Conclusions: We have provided treatment simulation and planning methods as well as guidelines for the evaluation of target coverage and normal tissue sparing that should facilitate the more aggressive treatment of cervical cancer. Ó 2008 Elsevier Inc. IMRT, Cervix cancer, PET/CT, Dose escalation, Treatment planning.
INTRODUCTION
mented a treatment technique for patients enrolled on a clinical trial investigating the use of position emission tomography (PET)/computed tomography (CT) imageguided IMRT for cervical carcinoma with PET-positive PALN (3, 4). The feasibility of using IMRT for irradiation of the pelvis has been examined by other investigators. Roeske et al. carried out a treatment planning study investigating the feasibility of pelvic irradiation using IMRT to improve smallbowel sparing for patients with cervical or endometrial cancer (5). The IMRT treatment plans were generated on the CT scan datasets of 10 patients. The clinical target volume (CTV) included the proximal vagina, parametrial tissues, uterus (if present), and regional lymph nodes. The planning target volume (PTV) was defined to be the CTV plus a 1.0-cm
Carcinoma of the cervix with positive para-aortic lymph nodes (PALN) has been treated historically using two distinct treatment volumes, pelvis and PALNs, using a combination of conventional external beam radiotherapy and brachytherapy. Given that the survival rate is approximately 30% at 5 years, more aggressive treatment is warranted for this patient population (1). A dose escalation study from the Radiation Therapy and Oncology Group (RTOG-92-10), using conventional external beam planning with chemotherapy for PALN irradiation, was found to lead to unacceptable bowel toxicity (2). Because intensity-modulated radiation therapy (IMRT) allows the delivery of escalated doses with sparing of surrounding normal tissue structures, we imple-
Conflict of interest: none. Received Sept 12, 2007, and in revised form Feb 18, 2008. Accepted for publication Feb 21, 2008.
Reprint requests to: Jacqueline Esthappan, Ph.D., Department of Radiation Oncology, Washington University School of Medicine, 4921 Parkview Place, Campus Box 8224, St. Louis, MO 63110. Tel: (314) 747-9859; Fax: (314) 747-9557; E-mail: esthappan@ radonc.wustl.edu 1134
PET/CT IMRT for cervical carcinoma d J. ESTHAPPAN et al.
margin. Four-field conventional plans were compared with four- and nine-field IMRT plans delivering 45.0 Gy to the PTV at 1.8 Gy per fraction. These investigators found that the percentage volume of small-bowel region (defined as the peritoneal cavity from the level of lumbar vertebral bodies L4 to L5 and excluding bladder and rectum) to receive greater than 30.0 Gy was reduced with IMRT. The percentage volumes of small-bowel region receiving 15.0 Gy in the conventional and IMRT plans were 25% and 13%, respectively (5). Roeske et al. concluded that IMRT to the pelvis provided an effective way to reduce the amount of irradiated small bowel, and that this method could potentially lead to fewer small-bowel complications (5). Mundt et al. continued this work by tracking acute gastrointestinal (GI) toxicity in cervical and endometrial cancer patients treated to 45.0 Gy at 1.8 Gy per fraction with a seven- or nine- field, 6-MV IMRT technique (6). In that study 40 IMRT patients were compared with 35 patients given conventional treatment with whole-pelvis irradiation. Target volume definition for the IMRT patients was similar to that described by Roeske et al. (5). All patients received low-dose-rate intracavitary brachytherapy after pelvic irradiation. No patient developed Grade 3 acute toxicity. There was a statistically significant difference in the rate of Grade 2 acute GI toxicity between IMRT-treated and conventionally treated patients (60% vs. 91%, respectively) (6). Subsequently, Mundt et al. tracked chronic GI toxicity in the same groups of patients (7). In that study, the IMRT-treated patients as compared with the conventionally treated patients had a statistically significant lower rate of chronic GI toxicity (11% vs. 50%, respectively). Median follow-up in the IMRT and conventionally treated groups were 19.6 and 30.2 months, respectively. Mundt et al. concluded that the IMRT treatment was well tolerated and led to less acute and chronic GI toxicities than use of whole-pelvis irradiation (6, 7). The feasibility of using IMRT for extended-field irradiation of the pelvis and PALNs has also been looked at by other investigators. In a dosimetric study, Portelance et al. investigated the feasibility of IMRT for irradiation of the PALNs as well as the pelvis for cervical cancer patients (8). Twoand four-field conventional treatment plans were compared with four-, seven-, and nine–field, 18-MV IMRT plans. The target volume included the para-aortic and iliac lymph nodes and the uterus. The amount of small bowel receiving 45.0 Gy was compared between the plans and found to be 11%, 15%, and 13.6% for the four-, seven-, and nine-field IMRT plans vs. approximately 35% for the conventional techniques (8). Portelance et al. concluded that for similar target coverage, sparing of normal tissue structures was better with IMRT (8). Salama et al. reported on their initial experience with extended-field IMRT irradiation to the para-aortic and pelvic nodes for 13 patients (9). The CTV in the pelvis was defined as described above (5). The CTV in the para-aortic region included the aorta and inferior vena cava with at least a 1.5cm margin, and, if necessary, modified further for kidney and bowel sparing. The prescription dose was 45.0 Gy at 1.8-Gy fractions daily, with an additional 9.0-Gy boost to gross disease in the para-aortics or pelvis. Two patients
1135
experienced Grade 3 or higher toxicity; median follow-up was 11 months. Salama et al. concluded that use of extended field IMRT was safe and associated with a low rate of acute toxicity (9). All of these studies demonstrated a benefit of IMRT for sparing of the bowel for cervical cancer patients; however, none of these studies involved the delivery of prescribed doses beyond 45 Gy. Our treatment technique is distinct from previously investigated IMRT techniques in that it is intended for dose escalation beyond 45.0 Gy. This technique has been implemented for patients enrolled on a clinical trial allowing dose escalation to 60.0 Gy to the positive lymph nodes and 50.0 Gy to the para-aortic and pelvic regions by means of PET/CT image– guided IMRT (10). The delivery of doses up to 60.0 Gy to the positive lymph nodes with concurrent sparing of the surrounding bowel and kidneys poses a challenge to the development of an acceptable IMRT treatment plan. Here, we review the isodose plans used for the treatment of 10 patients enrolled in the clinical trial. Based on this review, we describe methods for simulation and planning of these treatments. We also provide objectives for target coverage as well as normal tissue sparing to guide treatment plan evaluation. METHODS AND MATERIALS Image acquisition Patients were imaged supine on a multi-slice CT scanner equipped with isocenter localization software (Philips Medical Systems) to facilitate alpha cradle construction and isocenter localization. Before removing patients from the CT scanner, alignment marks were placed to facilitate repositioning of the patient in alpha cradle on a hybrid PET/CT scanner (Biograph LSO 2; Siemens Medical Solutions, Malvern, PA) for further imaging. Patients were intravenously administered 10-15 mCi (555 MBq) of 18F-fluorodeoxyglucose (FDG) and, 45 to 90 min later, underwent PET/CT scanning. For the CT component of the PET/CT, the scan parameters were 0.5-cm-thick CT images with 0.4-cm spacing from the middle of the neck to the upper thighs, 130 kVp, 111 mAs, spiral pitch of 1.0 to 1.3, and a 50-cm displayed field of view. For the PET component of the PET/CT, a series of four to six overlapping emission and transmission images were acquired over the same anatomical extent. Immediately after the PET/CT, a second CT scan of the patient was acquired using parameters similar to those stated above, except with 0.3-cm thickness and 0.25-cm spacing from the diaphragm to a few centimeters below the ischial tuberosities. This second CT scan was acquired for treatment planning and dose calculation purposes. The PET study and the second CT study were then transferred to a treatment planning workstation (Varian Eclipse Treatment Planning System V6.5, Varian Medical Systems, Palo Alto, CA). The two image studies were registered manually using pelvic and vertebral bony anatomy. In this manner, 10 patients were imaged.
Target volume definition On the CT images, the target volumes for the para-aortic and pelvic lymph node (PLN) bed were defined by contouring the appropriate vessel structures and adding margins (11). For the PALN bed target volume, the aorta and vena cava were contoured on the CT images from the renal vessels down to the aortic bifurcation. A margin of 0.7 cm was added to the contoured vessels, and the vertebral bodies were excluded to create a PALN bed clinical target volume
1136
I. J. Radiation Oncology d Biology d Physics
Volume 72, Number 4, 2008
MTVCERVIX, respectively. A single IMRT plan was used for concurrent treatment of the three target volumes. The IMRT plan used either seven megavoltage photon beams at 333, 282, 231, 180, 129, 78, and 27 degrees or 10 beams at 330, 290, 255, 225, 190, 144, 108, 72, 36, and 0 degrees (Varian convention scale). Beam energy was 6 MV, 18 MV, or a combination of the two energies. The IMRT treatment plans were optimized so that at least 95% of the PTVPLN+PALN and 95% of the MTVNODAL received approximately 95% of the prescribed doses. Minimum dose to the MTVCERVIX was to be at least 20.0 Gy. All patients received high-dose-rate intracavitary brachytherapy (39.0 Gy point A dose) (13).
Definition of organ-at-risk volumes For all patients, normal tissue structures contoured on the CT scan included the skin, bladder, rectum, femurs, kidneys, pelvic bones, vertebral bodies, and bowel. The small bowel and colon were contoured as a single structure labeled as ‘‘bowel.’’ Each vertebral body of the lumbar spine was contoured as a 2-cm diameter sphere centered inside each body, and then labeled as L1, L2, L3, L4, and L5. No margins were added to these contours for localization/ treatment set-up uncertainties.
Evaluation of dose to bowel, kidneys, and lumbar vertebral bodies
Fig. 1. Planning target volumes for one patient are shown above. Two target volumes are defined on the positron emission tomography (PET) images: metabolic target volume (MTV) for the PET-positive cervix (MTVCERVIX; red) and for the PET-positive para-aortic lymph nodes (PALN) and pelvic lymph nodes (both MTVNODAL; yellow). A composite target volume is defined on the CT images: PTVPLN+PALN for both the para-aortic and pelvic lymph node bed (shown in cyan). (CTVPALN_BED). For the PLN bed target volume the common, internal, and external iliac vessels were contoured on the CT images from the bifurcation of the aorta down to the medial circumflex femoral artery, and included the medial psoas muscle. A margin of 0.7 cm was added to the contoured vessels, and both the pelvic and vertebral bones were excluded to create a PLN clinical target volume (CTVPLN_BED). The CTVPLN_BED and CTVPALN_BED were summed together to create a composite CTV. A margin of 0.7 cm was added to the composite CTV to create a planning target volume (PTVPLN+PALN) for both the PALN and PLN bed. On the PET images, the PET-positive PALN and PLN and the cervix were defined by contouring the FDG-PET–enhancing volumes. The metabolic target volume (MTVNODAL) for the PETpositive PALN and PLN was defined on the PET images as those voxels with counts greater than 40% of the peak activity in the tumor (12). The metabolic target volume (MTVCERVIX) for the PET-positive cervix was defined similarly on the PET images. The metabolic target volumes with no additional margins served as surrogates for PTVs. All target volumes (MTVNODAL, MTVCERVIX, and PTVPLN+PALN) from one patient are shown as an example in Fig. 1.
Treatment planning Treatment planning and dose calculations were carried out on the CT images. Goal prescription doses of 60.0, 50.0, and 20.0 Gy in 30 fractions were assigned to the MTVNODAL, PTVPLN+PALN, and the
Once the treatment plans were generated and final dose calculated, the cumulative dose–volume histograms (DVHs) for the left kidney, right kidney, bowel (small bowel and colon together), and the lumbar spine were generated on the Eclipse treatment planning system (Varian), and then exported to MATLAB (The Mathworks, Inc., Natick, MA) for all 10 patients. A mean cumulative DVH for each structure was generated in MATLAB by averaging the volumes of the 10 DVHs at intervals of 0.10 Gy from 0.0 Gy to 70.0 Gy.
RESULTS The MTVNODAL and PTVPLN+PALN target coverage goals were satisfied in all plans. Minimum dose to the MTVCERVIX for all IMRT plans was at least 20.0 Gy. The mean DVH along with the DVHs for the individual patients are shown in Fig. 2 for bowel (small bowel plus colon). Analysis of mean DVHs indicated that treatment plans irradiated approximately 50% of the bowel volume to at least 25.0 Gy, with less than 10% receiving at least 50.0 Gy and less than 1% receiving at least 60.0 Gy. The mean DVH along with the DVHs for the individual patients for the right and left kidneys are shown in Figs. 3a and 3b, respectively. With regard to mean DVH for the right kidney, approximately 50% of the kidney volume received at least 16.0 Gy, less than 5% received at least 50.0 Gy, and less than 1% received at least 60.0 Gy. Similar results were obtained for the mean DVH for the left kidney. The spread of individual DVHs about the mean DVH were found to be similar for both kidneys, except for one patient with extensive gross disease, involving the delivery of high doses through the left kidney and hence poor sparing of that kidney (Fig. 3b). The mean DVH along with the DVHs for the individual patients for the vertebral bodies of the lumbar spine, L1 to L5, are depicted in Figs. 4a to 4e, respectively. The mean DVHs (Figs. 4a–4e) for the vertebral bodies in the lumbar
PET/CT IMRT for cervical carcinoma d J. ESTHAPPAN et al.
Fig. 2. Mean dose–volume histogram (DVH; bold) and individual DVHs (solid lines) for bowel (i.e., small bowel plus colon).
spine show that 50% of the contoured volume in each vertebral body receives between 44.0 and 48.0 Gy. The individual DVHs show a wide spread about the mean DVH, and is most likely because the location of the PET-positive PALN relative to the lumbar vertebral bodies varied between patients. DISCUSSION Analysis of the mean and individual DVHs enhanced our understanding of the results that can be expected, in terms of bowel and kidney sparing, when trying to deliver 60.0 Gy to the PET-positive para-aortic and pelvic lymph nodes and 50.0 Gy to the para-aortic and pelvic lymph node bed. We present the mean DVHs as objectives that can be used to evaluate the quality of an IMRT treatment plan that uses the treatment techniques described here. The dose to bowel and kidneys should be kept as low as possible during the development of a treatment plan. The mean bowel DVH presented here can be used as a priori information to help guide treatment plan optimization. Our results indicate that it is possible to achieve plans in which
1137
the volume of bowel receiving at least 45.0 Gy is less than 15%, and this may be a reasonable starting point for the optimization. Similarly, the mean kidney DVHs can be used to guide treatment plan optimization, although the extent of disease present for each individual patient should also be taken into consideration. As seen in Fig. 3b, deviations from the mean kidney DVH can occur for patients with more extensive gross disease, in whom the target may be located in close proximity to one or both of the kidneys. Our IMRT treatment planning techniques did not include optimization of dose to the lumbar spine. However in the future it may be worthwhile to investigate the ability of IMRT to spare the lumbar spine. This is because many of the patients receiving radiation also undergo chemotherapy, leading to depletion of bone marrow cells. Reduction of external beam radiation dose to the bone marrow through IMRT techniques may be of benefit to patients receiving chemotherapy. The use of IMRT for bone marrow sparing has been investigated previously and has been shown to be feasible and to have a favorable impact on hematologic toxicity (14, 15). However these studies focused on doses of 45.0 Gy to the pelvic region and analyzed hematologic toxicities of bone marrow in L4 to L5, sacrum, and medial iliac crest. The treatment planning technique presented here involves the delivery of 50.0 Gy and 60.0 Gy to the PALN region, which is directly adjacent and in close proximity to the entire lumbar spine. For this reason, sparing of the lumbar spine with concurrent irradiation of the PALN region to higher dose levels may be difficult to achieve by use of IMRT. With the data that we have accumulated using our treatment techniques, we plan to correlate the clinical outcomes of these patients with their DVHs to determine whether there is a need for dose optimization to the lumbar spine. CONCLUSION We have outlined a treatment technique that uses IMRT to deliver doses as high as 60.0 Gy for patients with cervical cancer and positive PALN. As part of this technique, we have provided a description of the image acquisition
Fig. 3. Mean dose–volume histogram (DVH; bold) and individual DVHs (solid lines) for (a) right kidney and (b) left kidney.
1138
I. J. Radiation Oncology d Biology d Physics
Volume 72, Number 4, 2008
Fig. 4. Mean dose–volume histogram (DVH; bold) and individual DVHs (solid lines) for a 2-cm diameter sphere contoured inside the center of (a) L1, (b) L2, (c) L3, (d) L4, and (e) L5.
technique using a multi-slice PET/CT scanner. While CT provides adequate resolution for definition of the nodal target volumes, FDG-PET imaging highlights metabolically active tumor volumes that can be treated to escalated doses. We have described the definition of target volumes based on PET/CT imaging and the doses that are prescribed to these different target volumes. We have reviewed the isodose plans of 10 patients treated with the technique described here. With
IMRT we were able to deliver at least 95% of the prescribed doses to 95% of the target volumes for all 10 patients. With this level of coverage, the amount of sparing that we were able to achieve is described in the DVH-based data that we provide here. The resulting DVHs for the bowel and kidneys can be used as objectives for treatment planning. The treatment planning methods described here should facilitate more aggressive treatment of cervical cancer.
REFERENCES 1. Rotman M, Pajak TF, Choi K, et al. Prophylactic extended-field irradiation of para-aortic lymph nodes in stages IIB and bulky IB and IIA cervical carcinomas. Ten-year treatment results of RTOG 79-20. JAMA 1995;274:387–393.
2. Grigsby PW, Heydon K, Mutch DG, et al. Long-term followup of RTOG 92-10: Cervical cancer with positive para-aortic lymph nodes. Int J Radiat Oncol Biol Phys 2001;51: 982–987.
PET/CT IMRT for cervical carcinoma d J. ESTHAPPAN et al.
3. Esthappan J, Mutic S, Malyapa RS, et al. Treatment planning guidelines regarding the use of CT/PET-guided IMRT for cervical carcinoma with positive paraaortic lymph nodes. Int J Radiat Oncol Biol Phys 2004;58:1289–1297. 4. Mutic S, Malyapa RS, Grigsby PW, et al. PET-guided IMRT for cervical carcinoma with positive para-aortic lymph nodes— a dose-escalation treatment planning study. Int J Radiat Oncol Biol Phys 2003;55:28–35. 5. Roeske JC, Lujan A, Rotmensch J, et al. Intensity-modulated whole pelvic radiation therapy in patients with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2000;48:1613–1621. 6. Mundt AJ, Lujan AE, Rotmensch J, et al. Intensity-modulated whole pelvic radiotherapy in women with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2002;52:1330–1337. 7. Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 2003;56:1354–1360. 8. Portelance L, Chao KS, Grigsby PW, et al. Intensity-modulated radiation therapy (IMRT) reduces small bowel, rectum, and bladder doses in patients with cervical cancer receiving pelvic and para-aortic irradiation. Int J Radiat Oncol Biol Phys 2001; 51:261–266. 9. Salama JK, Mundt AJ, Roeske J, et al. Preliminary outcome and toxicity report of extended-field, intensity-modulated radiation
10.
11.
12.
13.
14.
15.
1139
therapy for gynecologic malignancies. Int J Radiat Oncol Biol Phys 2006;65:1170–1176. Singh K, Lale A, Eong Ooi E, et al. A prospective clinical study on the use of reverse transcription-polymerase chain reaction for the early diagnosis of dengue fever. J Mol Diagn 2006;8: 613–616; quiz 617–620. Taylor A, Rockall AG, Reznek RH, et al. Mapping pelvic lymph nodes: Guidelines for delineation in intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2005;63:1604–1612. Miller TR, Grigsby PW. Measurement of tumor volume by PET to evaluate prognosis in patients with advanced cervical cancer treated by radiation therapy. Int J Radiat Oncol Biol Phys 2002; 53:353–359. Grigsby PW, Perez CA, Chao KS, et al. Radiation therapy for carcinoma of the cervix with biopsy-proven positive para-aortic lymph nodes. Int J Radiat Oncol Biol Phys 2001;49:733–738. Brixey CJ, Roeske JC, Lujan AE, et al. Impact of intensitymodulated radiotherapy on acute hematologic toxicity in women with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2002;54:1388–1396. Roeske JC, Lujan A, Reba RC, et al. Incorporation of SPECT bone marrow imaging into intensity modulated whole-pelvic radiation therapy treatment planning for gynecologic malignancies. Radiother Oncol 2005;77:11–17.
doi:10.1016/j.ijrobp.2008.02.063
CLINICAL INVESTIGATION
Cervix
PROSPECTIVE CLINICAL TRIAL OF POSITRON EMISSION TOMOGRAPHY/ COMPUTED TOMOGRAPHY IMAGE-GUIDED INTENSITY-MODULATED RADIATION THERAPY FOR CERVICAL CARCINOMA WITH POSITIVE PARA-AORTIC LYMPH NODES JACQUELINE ESTHAPPAN, PH.D.,* SUMMER CHAUDHARI, M.S.,* LAKSHMI SANTANAM, PH.D,* SASA MUTIC, M.S.,* JEFFREY OLSEN, B.S.,* DUSTEN M. MACDONALD, M.D.,* DANIEL A. LOW, PH.D.,* ANURAG K. SINGH, M.D.,y AND PERRY W. GRIGSBY, M.D.* * Department of Radiation Oncology, Washington University School of Medicine, St. Louis, MO; and y Radiation Oncology Branch, National Cancer Institute, Bethesda, MD Purpose: To describe a more aggressive treatment technique allowing dose escalation to positive para-aortic lymph nodes (PALN) in patients with cervical cancer, by means of positron emission tomography (PET)/computed tomography (CT)–guided intensity-modulated radiation therapy (IMRT). Here, we describe methods for simulation and planning of these treatments and provide objectives for target coverage as well as normal tissue sparing to guide treatment plan evaluation. Methods and Materials: Patients underwent simulation on a PET/CT scanner. Treatment plans were generated to deliver 60.0 Gy to the PET-positive PALN and 50.0 Gy to the PALN and pelvic lymph node beds. Treatment plans were optimized to deliver at least 95% of the prescribed doses to at least 95% of each target volume. Dose–volume histograms were calculated for normal structures. Results: The plans of 10 patients were reviewed. Target coverage goals were satisfied in all plans. Analysis of dose– volume histograms indicated that treatment plans involved irradiation of approximately 50% of the bowel volume to at least 25.0 Gy, with less than 10% receiving at least 50.0 Gy and less than 1% receiving at least 60.0. With regard to kidney sparing, approximately 50% of the kidney volume received at least 16.0 Gy, less than 5% received at least 50.0 Gy, and less than 1% received at least 60.0 Gy. Conclusions: We have provided treatment simulation and planning methods as well as guidelines for the evaluation of target coverage and normal tissue sparing that should facilitate the more aggressive treatment of cervical cancer. Ó 2008 Elsevier Inc. IMRT, Cervix cancer, PET/CT, Dose escalation, Treatment planning.
INTRODUCTION
mented a treatment technique for patients enrolled on a clinical trial investigating the use of position emission tomography (PET)/computed tomography (CT) imageguided IMRT for cervical carcinoma with PET-positive PALN (3, 4). The feasibility of using IMRT for irradiation of the pelvis has been examined by other investigators. Roeske et al. carried out a treatment planning study investigating the feasibility of pelvic irradiation using IMRT to improve smallbowel sparing for patients with cervical or endometrial cancer (5). The IMRT treatment plans were generated on the CT scan datasets of 10 patients. The clinical target volume (CTV) included the proximal vagina, parametrial tissues, uterus (if present), and regional lymph nodes. The planning target volume (PTV) was defined to be the CTV plus a 1.0-cm
Carcinoma of the cervix with positive para-aortic lymph nodes (PALN) has been treated historically using two distinct treatment volumes, pelvis and PALNs, using a combination of conventional external beam radiotherapy and brachytherapy. Given that the survival rate is approximately 30% at 5 years, more aggressive treatment is warranted for this patient population (1). A dose escalation study from the Radiation Therapy and Oncology Group (RTOG-92-10), using conventional external beam planning with chemotherapy for PALN irradiation, was found to lead to unacceptable bowel toxicity (2). Because intensity-modulated radiation therapy (IMRT) allows the delivery of escalated doses with sparing of surrounding normal tissue structures, we imple-
Conflict of interest: none. Received Sept 12, 2007, and in revised form Feb 18, 2008. Accepted for publication Feb 21, 2008.
Reprint requests to: Jacqueline Esthappan, Ph.D., Department of Radiation Oncology, Washington University School of Medicine, 4921 Parkview Place, Campus Box 8224, St. Louis, MO 63110. Tel: (314) 747-9859; Fax: (314) 747-9557; E-mail: esthappan@ radonc.wustl.edu 1134
PET/CT IMRT for cervical carcinoma d J. ESTHAPPAN et al.
margin. Four-field conventional plans were compared with four- and nine-field IMRT plans delivering 45.0 Gy to the PTV at 1.8 Gy per fraction. These investigators found that the percentage volume of small-bowel region (defined as the peritoneal cavity from the level of lumbar vertebral bodies L4 to L5 and excluding bladder and rectum) to receive greater than 30.0 Gy was reduced with IMRT. The percentage volumes of small-bowel region receiving 15.0 Gy in the conventional and IMRT plans were 25% and 13%, respectively (5). Roeske et al. concluded that IMRT to the pelvis provided an effective way to reduce the amount of irradiated small bowel, and that this method could potentially lead to fewer small-bowel complications (5). Mundt et al. continued this work by tracking acute gastrointestinal (GI) toxicity in cervical and endometrial cancer patients treated to 45.0 Gy at 1.8 Gy per fraction with a seven- or nine- field, 6-MV IMRT technique (6). In that study 40 IMRT patients were compared with 35 patients given conventional treatment with whole-pelvis irradiation. Target volume definition for the IMRT patients was similar to that described by Roeske et al. (5). All patients received low-dose-rate intracavitary brachytherapy after pelvic irradiation. No patient developed Grade 3 acute toxicity. There was a statistically significant difference in the rate of Grade 2 acute GI toxicity between IMRT-treated and conventionally treated patients (60% vs. 91%, respectively) (6). Subsequently, Mundt et al. tracked chronic GI toxicity in the same groups of patients (7). In that study, the IMRT-treated patients as compared with the conventionally treated patients had a statistically significant lower rate of chronic GI toxicity (11% vs. 50%, respectively). Median follow-up in the IMRT and conventionally treated groups were 19.6 and 30.2 months, respectively. Mundt et al. concluded that the IMRT treatment was well tolerated and led to less acute and chronic GI toxicities than use of whole-pelvis irradiation (6, 7). The feasibility of using IMRT for extended-field irradiation of the pelvis and PALNs has also been looked at by other investigators. In a dosimetric study, Portelance et al. investigated the feasibility of IMRT for irradiation of the PALNs as well as the pelvis for cervical cancer patients (8). Twoand four-field conventional treatment plans were compared with four-, seven-, and nine–field, 18-MV IMRT plans. The target volume included the para-aortic and iliac lymph nodes and the uterus. The amount of small bowel receiving 45.0 Gy was compared between the plans and found to be 11%, 15%, and 13.6% for the four-, seven-, and nine-field IMRT plans vs. approximately 35% for the conventional techniques (8). Portelance et al. concluded that for similar target coverage, sparing of normal tissue structures was better with IMRT (8). Salama et al. reported on their initial experience with extended-field IMRT irradiation to the para-aortic and pelvic nodes for 13 patients (9). The CTV in the pelvis was defined as described above (5). The CTV in the para-aortic region included the aorta and inferior vena cava with at least a 1.5cm margin, and, if necessary, modified further for kidney and bowel sparing. The prescription dose was 45.0 Gy at 1.8-Gy fractions daily, with an additional 9.0-Gy boost to gross disease in the para-aortics or pelvis. Two patients
1135
experienced Grade 3 or higher toxicity; median follow-up was 11 months. Salama et al. concluded that use of extended field IMRT was safe and associated with a low rate of acute toxicity (9). All of these studies demonstrated a benefit of IMRT for sparing of the bowel for cervical cancer patients; however, none of these studies involved the delivery of prescribed doses beyond 45 Gy. Our treatment technique is distinct from previously investigated IMRT techniques in that it is intended for dose escalation beyond 45.0 Gy. This technique has been implemented for patients enrolled on a clinical trial allowing dose escalation to 60.0 Gy to the positive lymph nodes and 50.0 Gy to the para-aortic and pelvic regions by means of PET/CT image– guided IMRT (10). The delivery of doses up to 60.0 Gy to the positive lymph nodes with concurrent sparing of the surrounding bowel and kidneys poses a challenge to the development of an acceptable IMRT treatment plan. Here, we review the isodose plans used for the treatment of 10 patients enrolled in the clinical trial. Based on this review, we describe methods for simulation and planning of these treatments. We also provide objectives for target coverage as well as normal tissue sparing to guide treatment plan evaluation. METHODS AND MATERIALS Image acquisition Patients were imaged supine on a multi-slice CT scanner equipped with isocenter localization software (Philips Medical Systems) to facilitate alpha cradle construction and isocenter localization. Before removing patients from the CT scanner, alignment marks were placed to facilitate repositioning of the patient in alpha cradle on a hybrid PET/CT scanner (Biograph LSO 2; Siemens Medical Solutions, Malvern, PA) for further imaging. Patients were intravenously administered 10-15 mCi (555 MBq) of 18F-fluorodeoxyglucose (FDG) and, 45 to 90 min later, underwent PET/CT scanning. For the CT component of the PET/CT, the scan parameters were 0.5-cm-thick CT images with 0.4-cm spacing from the middle of the neck to the upper thighs, 130 kVp, 111 mAs, spiral pitch of 1.0 to 1.3, and a 50-cm displayed field of view. For the PET component of the PET/CT, a series of four to six overlapping emission and transmission images were acquired over the same anatomical extent. Immediately after the PET/CT, a second CT scan of the patient was acquired using parameters similar to those stated above, except with 0.3-cm thickness and 0.25-cm spacing from the diaphragm to a few centimeters below the ischial tuberosities. This second CT scan was acquired for treatment planning and dose calculation purposes. The PET study and the second CT study were then transferred to a treatment planning workstation (Varian Eclipse Treatment Planning System V6.5, Varian Medical Systems, Palo Alto, CA). The two image studies were registered manually using pelvic and vertebral bony anatomy. In this manner, 10 patients were imaged.
Target volume definition On the CT images, the target volumes for the para-aortic and pelvic lymph node (PLN) bed were defined by contouring the appropriate vessel structures and adding margins (11). For the PALN bed target volume, the aorta and vena cava were contoured on the CT images from the renal vessels down to the aortic bifurcation. A margin of 0.7 cm was added to the contoured vessels, and the vertebral bodies were excluded to create a PALN bed clinical target volume
1136
I. J. Radiation Oncology d Biology d Physics
Volume 72, Number 4, 2008
MTVCERVIX, respectively. A single IMRT plan was used for concurrent treatment of the three target volumes. The IMRT plan used either seven megavoltage photon beams at 333, 282, 231, 180, 129, 78, and 27 degrees or 10 beams at 330, 290, 255, 225, 190, 144, 108, 72, 36, and 0 degrees (Varian convention scale). Beam energy was 6 MV, 18 MV, or a combination of the two energies. The IMRT treatment plans were optimized so that at least 95% of the PTVPLN+PALN and 95% of the MTVNODAL received approximately 95% of the prescribed doses. Minimum dose to the MTVCERVIX was to be at least 20.0 Gy. All patients received high-dose-rate intracavitary brachytherapy (39.0 Gy point A dose) (13).
Definition of organ-at-risk volumes For all patients, normal tissue structures contoured on the CT scan included the skin, bladder, rectum, femurs, kidneys, pelvic bones, vertebral bodies, and bowel. The small bowel and colon were contoured as a single structure labeled as ‘‘bowel.’’ Each vertebral body of the lumbar spine was contoured as a 2-cm diameter sphere centered inside each body, and then labeled as L1, L2, L3, L4, and L5. No margins were added to these contours for localization/ treatment set-up uncertainties.
Evaluation of dose to bowel, kidneys, and lumbar vertebral bodies
Fig. 1. Planning target volumes for one patient are shown above. Two target volumes are defined on the positron emission tomography (PET) images: metabolic target volume (MTV) for the PET-positive cervix (MTVCERVIX; red) and for the PET-positive para-aortic lymph nodes (PALN) and pelvic lymph nodes (both MTVNODAL; yellow). A composite target volume is defined on the CT images: PTVPLN+PALN for both the para-aortic and pelvic lymph node bed (shown in cyan). (CTVPALN_BED). For the PLN bed target volume the common, internal, and external iliac vessels were contoured on the CT images from the bifurcation of the aorta down to the medial circumflex femoral artery, and included the medial psoas muscle. A margin of 0.7 cm was added to the contoured vessels, and both the pelvic and vertebral bones were excluded to create a PLN clinical target volume (CTVPLN_BED). The CTVPLN_BED and CTVPALN_BED were summed together to create a composite CTV. A margin of 0.7 cm was added to the composite CTV to create a planning target volume (PTVPLN+PALN) for both the PALN and PLN bed. On the PET images, the PET-positive PALN and PLN and the cervix were defined by contouring the FDG-PET–enhancing volumes. The metabolic target volume (MTVNODAL) for the PETpositive PALN and PLN was defined on the PET images as those voxels with counts greater than 40% of the peak activity in the tumor (12). The metabolic target volume (MTVCERVIX) for the PET-positive cervix was defined similarly on the PET images. The metabolic target volumes with no additional margins served as surrogates for PTVs. All target volumes (MTVNODAL, MTVCERVIX, and PTVPLN+PALN) from one patient are shown as an example in Fig. 1.
Treatment planning Treatment planning and dose calculations were carried out on the CT images. Goal prescription doses of 60.0, 50.0, and 20.0 Gy in 30 fractions were assigned to the MTVNODAL, PTVPLN+PALN, and the
Once the treatment plans were generated and final dose calculated, the cumulative dose–volume histograms (DVHs) for the left kidney, right kidney, bowel (small bowel and colon together), and the lumbar spine were generated on the Eclipse treatment planning system (Varian), and then exported to MATLAB (The Mathworks, Inc., Natick, MA) for all 10 patients. A mean cumulative DVH for each structure was generated in MATLAB by averaging the volumes of the 10 DVHs at intervals of 0.10 Gy from 0.0 Gy to 70.0 Gy.
RESULTS The MTVNODAL and PTVPLN+PALN target coverage goals were satisfied in all plans. Minimum dose to the MTVCERVIX for all IMRT plans was at least 20.0 Gy. The mean DVH along with the DVHs for the individual patients are shown in Fig. 2 for bowel (small bowel plus colon). Analysis of mean DVHs indicated that treatment plans irradiated approximately 50% of the bowel volume to at least 25.0 Gy, with less than 10% receiving at least 50.0 Gy and less than 1% receiving at least 60.0 Gy. The mean DVH along with the DVHs for the individual patients for the right and left kidneys are shown in Figs. 3a and 3b, respectively. With regard to mean DVH for the right kidney, approximately 50% of the kidney volume received at least 16.0 Gy, less than 5% received at least 50.0 Gy, and less than 1% received at least 60.0 Gy. Similar results were obtained for the mean DVH for the left kidney. The spread of individual DVHs about the mean DVH were found to be similar for both kidneys, except for one patient with extensive gross disease, involving the delivery of high doses through the left kidney and hence poor sparing of that kidney (Fig. 3b). The mean DVH along with the DVHs for the individual patients for the vertebral bodies of the lumbar spine, L1 to L5, are depicted in Figs. 4a to 4e, respectively. The mean DVHs (Figs. 4a–4e) for the vertebral bodies in the lumbar
PET/CT IMRT for cervical carcinoma d J. ESTHAPPAN et al.
Fig. 2. Mean dose–volume histogram (DVH; bold) and individual DVHs (solid lines) for bowel (i.e., small bowel plus colon).
spine show that 50% of the contoured volume in each vertebral body receives between 44.0 and 48.0 Gy. The individual DVHs show a wide spread about the mean DVH, and is most likely because the location of the PET-positive PALN relative to the lumbar vertebral bodies varied between patients. DISCUSSION Analysis of the mean and individual DVHs enhanced our understanding of the results that can be expected, in terms of bowel and kidney sparing, when trying to deliver 60.0 Gy to the PET-positive para-aortic and pelvic lymph nodes and 50.0 Gy to the para-aortic and pelvic lymph node bed. We present the mean DVHs as objectives that can be used to evaluate the quality of an IMRT treatment plan that uses the treatment techniques described here. The dose to bowel and kidneys should be kept as low as possible during the development of a treatment plan. The mean bowel DVH presented here can be used as a priori information to help guide treatment plan optimization. Our results indicate that it is possible to achieve plans in which
1137
the volume of bowel receiving at least 45.0 Gy is less than 15%, and this may be a reasonable starting point for the optimization. Similarly, the mean kidney DVHs can be used to guide treatment plan optimization, although the extent of disease present for each individual patient should also be taken into consideration. As seen in Fig. 3b, deviations from the mean kidney DVH can occur for patients with more extensive gross disease, in whom the target may be located in close proximity to one or both of the kidneys. Our IMRT treatment planning techniques did not include optimization of dose to the lumbar spine. However in the future it may be worthwhile to investigate the ability of IMRT to spare the lumbar spine. This is because many of the patients receiving radiation also undergo chemotherapy, leading to depletion of bone marrow cells. Reduction of external beam radiation dose to the bone marrow through IMRT techniques may be of benefit to patients receiving chemotherapy. The use of IMRT for bone marrow sparing has been investigated previously and has been shown to be feasible and to have a favorable impact on hematologic toxicity (14, 15). However these studies focused on doses of 45.0 Gy to the pelvic region and analyzed hematologic toxicities of bone marrow in L4 to L5, sacrum, and medial iliac crest. The treatment planning technique presented here involves the delivery of 50.0 Gy and 60.0 Gy to the PALN region, which is directly adjacent and in close proximity to the entire lumbar spine. For this reason, sparing of the lumbar spine with concurrent irradiation of the PALN region to higher dose levels may be difficult to achieve by use of IMRT. With the data that we have accumulated using our treatment techniques, we plan to correlate the clinical outcomes of these patients with their DVHs to determine whether there is a need for dose optimization to the lumbar spine. CONCLUSION We have outlined a treatment technique that uses IMRT to deliver doses as high as 60.0 Gy for patients with cervical cancer and positive PALN. As part of this technique, we have provided a description of the image acquisition
Fig. 3. Mean dose–volume histogram (DVH; bold) and individual DVHs (solid lines) for (a) right kidney and (b) left kidney.
1138
I. J. Radiation Oncology d Biology d Physics
Volume 72, Number 4, 2008
Fig. 4. Mean dose–volume histogram (DVH; bold) and individual DVHs (solid lines) for a 2-cm diameter sphere contoured inside the center of (a) L1, (b) L2, (c) L3, (d) L4, and (e) L5.
technique using a multi-slice PET/CT scanner. While CT provides adequate resolution for definition of the nodal target volumes, FDG-PET imaging highlights metabolically active tumor volumes that can be treated to escalated doses. We have described the definition of target volumes based on PET/CT imaging and the doses that are prescribed to these different target volumes. We have reviewed the isodose plans of 10 patients treated with the technique described here. With
IMRT we were able to deliver at least 95% of the prescribed doses to 95% of the target volumes for all 10 patients. With this level of coverage, the amount of sparing that we were able to achieve is described in the DVH-based data that we provide here. The resulting DVHs for the bowel and kidneys can be used as objectives for treatment planning. The treatment planning methods described here should facilitate more aggressive treatment of cervical cancer.
REFERENCES 1. Rotman M, Pajak TF, Choi K, et al. Prophylactic extended-field irradiation of para-aortic lymph nodes in stages IIB and bulky IB and IIA cervical carcinomas. Ten-year treatment results of RTOG 79-20. JAMA 1995;274:387–393.
2. Grigsby PW, Heydon K, Mutch DG, et al. Long-term followup of RTOG 92-10: Cervical cancer with positive para-aortic lymph nodes. Int J Radiat Oncol Biol Phys 2001;51: 982–987.
PET/CT IMRT for cervical carcinoma d J. ESTHAPPAN et al.
3. Esthappan J, Mutic S, Malyapa RS, et al. Treatment planning guidelines regarding the use of CT/PET-guided IMRT for cervical carcinoma with positive paraaortic lymph nodes. Int J Radiat Oncol Biol Phys 2004;58:1289–1297. 4. Mutic S, Malyapa RS, Grigsby PW, et al. PET-guided IMRT for cervical carcinoma with positive para-aortic lymph nodes— a dose-escalation treatment planning study. Int J Radiat Oncol Biol Phys 2003;55:28–35. 5. Roeske JC, Lujan A, Rotmensch J, et al. Intensity-modulated whole pelvic radiation therapy in patients with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2000;48:1613–1621. 6. Mundt AJ, Lujan AE, Rotmensch J, et al. Intensity-modulated whole pelvic radiotherapy in women with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2002;52:1330–1337. 7. Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with intensity-modulated whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys 2003;56:1354–1360. 8. Portelance L, Chao KS, Grigsby PW, et al. Intensity-modulated radiation therapy (IMRT) reduces small bowel, rectum, and bladder doses in patients with cervical cancer receiving pelvic and para-aortic irradiation. Int J Radiat Oncol Biol Phys 2001; 51:261–266. 9. Salama JK, Mundt AJ, Roeske J, et al. Preliminary outcome and toxicity report of extended-field, intensity-modulated radiation
10.
11.
12.
13.
14.
15.
1139
therapy for gynecologic malignancies. Int J Radiat Oncol Biol Phys 2006;65:1170–1176. Singh K, Lale A, Eong Ooi E, et al. A prospective clinical study on the use of reverse transcription-polymerase chain reaction for the early diagnosis of dengue fever. J Mol Diagn 2006;8: 613–616; quiz 617–620. Taylor A, Rockall AG, Reznek RH, et al. Mapping pelvic lymph nodes: Guidelines for delineation in intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2005;63:1604–1612. Miller TR, Grigsby PW. Measurement of tumor volume by PET to evaluate prognosis in patients with advanced cervical cancer treated by radiation therapy. Int J Radiat Oncol Biol Phys 2002; 53:353–359. Grigsby PW, Perez CA, Chao KS, et al. Radiation therapy for carcinoma of the cervix with biopsy-proven positive para-aortic lymph nodes. Int J Radiat Oncol Biol Phys 2001;49:733–738. Brixey CJ, Roeske JC, Lujan AE, et al. Impact of intensitymodulated radiotherapy on acute hematologic toxicity in women with gynecologic malignancies. Int J Radiat Oncol Biol Phys 2002;54:1388–1396. Roeske JC, Lujan A, Reba RC, et al. Incorporation of SPECT bone marrow imaging into intensity modulated whole-pelvic radiation therapy treatment planning for gynecologic malignancies. Radiother Oncol 2005;77:11–17.