1. Harvard Medical School Phase II Study of Proton Radiation Therapy for Spine and Paraspinal Sarcomas Thomas F. DeLaney, Norbert J. Liebsch, Ira J. Spiro, Thomas F. DeLaney, Norbert J. Liebsch, Ira J. Spiro, Patricia L. McManus, Judith Adams, Susan Dean, Patricia L. McManus, Judith Adams, Susan Dean, Francis J. Hornicek, Francis X. Pedlow, Francis J. Hornicek, Francis X. Pedlow, Andrew L. Rosenberg, G. Petur Nielsen, Andrew L. Rosenberg, G. Petur Nielsen, David C. Harmon, Sam S. Yoon, David C. Harmon, Sam S. Yoon, Kevin A. Raskin, Herman D. Suit

2. Harvard Medical School Sarcomas of the Spine Low rates of control with standard therapy Surgical margins are often positive because of proximity of critical normal tissues – –Radiotherapy doses need to be > 66 Gy Difficult to deliver with spinal cord dose constraint of ~ 50 Gy Chordoma (Sacrum) – –44% crude local failure rate (Mayo Clinic) Chondrosarcoma (Spine/Sacrum) – –42% crude local failure rate (Göteborg, Sweden) Osteosarcoma – –15/22 (68%) local failure (Ozaki, 2002) – –Contrast with extremity where local failures < 10%

3. Harvard Medical School Sarcomas of the Spine Hug et al. (Harvard Cyclotron Laboratory) – –Photon/proton XRT +/- resection1980-1992 – –47 pts: osteo-/chondrogenic axial skeletal tumors – –XRT: Post-op (23), Pre/post-op (17), XRT only (7) Group I: Chordoma/Chondrosarcoma n=20 – –Mean 73.9 CGE5 Year LC53%/100% Group II: Osteosarcoma n=15 – –Mean69.8 CGE5 Year LC59% Group III: Giant cell tumor, osteo or chondroblastoma n=12 – –Mean61.8 CGE 5 Year LC83% – –Trend to improved LC: primary, > 77 Gy, less residual

4. Harvard Medical School Spine and Paraspinal Sarcoma Spine/Paraspinal Sarcoma (PI T. DeLaney MD)Spine/Paraspinal Sarcoma (PI T. DeLaney MD) –Surgery + IORT ( Dura Plaque) + Photon/Proton IRB-approved protocolIRB-approved protocol Surgery: Maximal debulking/spine stabilizationSurgery: Maximal debulking/spine stabilization IORT: 90 Y dural plaque: 10 GyIORT: 90 Y dural plaque: 10 Gy Photon/Proton RadiotherapyPhoton/Proton Radiotherapy –70.2 CGE (Microscopic residual)* –77.4 CGE (Gross residual disease)* * Doses modified for concurrent chemotherapy, diabetes, connective tissue diseases, radioresponsive histologies

5. Harvard Medical School Ideal Dose Distribution

6. Harvard Medical School L1 Angiosarcoma ProtonIMRT ASTRO 2003

7. Harvard Medical School Paravertebral Epithelioid Sarcoma Intensity Modulated Protons (IMPT) vs. Intensity Modulated Photons (IMRT) (7 field) IMPTIMXT

8. Harvard Medical School Spine/Paraspinal Sarcoma: Clinical Trial Radiation TherapyRadiation Therapy CTV1 (clinical target volume)CTV1 (clinical target volume) –Tissues suspected of microscopic involvement –Doses: 50.4 CGE at 1.8 Gy per fraction Could be given with photonsCould be given with photons Majority of pts received ~30.6 Gy with photons (conformal or IMRT)Majority of pts received ~30.6 Gy with photons (conformal or IMRT) –Treated pre-operatively when possible to doses of: –19.8 Gy : Sacrum45-50.4 Gy: Thoracolumbar spine Reduces risk of tumor seeding into woundReduces risk of tumor seeding into wound Dose selection based upon risk of wound healing delayDose selection based upon risk of wound healing delay CTV dose reduced to 46.8 Gy with IDDM/autoimmune diseaseCTV dose reduced to 46.8 Gy with IDDM/autoimmune disease

9. Harvard Medical School Spine/Paraspinal Sarcoma: Clinical Trial Radiation TherapyRadiation Therapy –CTV (clinical target volume) Controlled information on extent of CTV not availableControlled information on extent of CTV not available CTV based upon the patterns of failure in surgical and radiation therapy seriesCTV based upon the patterns of failure in surgical and radiation therapy series Generally included the entire involved vertebral body and at least hemi-sacral segment in the CTVGenerally included the entire involved vertebral body and at least hemi-sacral segment in the CTV Areas of extra-osseous extension into soft tissue are included with 1 cm margin or to fascial barrierAreas of extra-osseous extension into soft tissue are included with 1 cm margin or to fascial barrier Scar coverage for patients receiving post-op XRT dependent upon histology, grade, length of scarScar coverage for patients receiving post-op XRT dependent upon histology, grade, length of scar

10. Harvard Medical School Spine/Paraspinal Sarcoma: Clinical Trial Radiation TherapyRadiation Therapy GTV (gross tumor volume)GTV (gross tumor volume) –CTV2: Encompasses original gross tumor Boosted to dose of 70.2 CGE at 1.8 CGE q.d. with protonsBoosted to dose of 70.2 CGE at 1.8 CGE q.d. with protons Giant cell tumor, Ewing’s sarcoma : 61.2 GyGiant cell tumor, Ewing’s sarcoma : 61.2 Gy –CTV3: Residual gross disease after surgery Boosted to dose of 77.4 CGE at 1.8 CGE q.d. with protonsBoosted to dose of 77.4 CGE at 1.8 CGE q.d. with protons If IDDM/autoimmune disease or concurrent chemotherapy (i.e. osteosarcoma), doses reduced by 8%: total 70.2-72.0 CGEIf IDDM/autoimmune disease or concurrent chemotherapy (i.e. osteosarcoma), doses reduced by 8%: total 70.2-72.0 CGE ChemotherapyChemotherapy –Adriamycin not delivered concurrently

11. Harvard Medical School Spine/Paraspinal Sarcoma: CT myelogram for Radiation Planning T6 spine/paraspinal Chondrosarcoma CT myelogram for lesions above the conus to outline spinal cord

12. Harvard Medical School Spine/Paraspinal Sarcoma: Treatment Position LPO Proton Field Thoracic Chondrosarcoma

13. Harvard Medical School Spine/Paraspinal Sarcoma: Treatment Planning S1 Chondrosarcoma Treatment Plan

14. Harvard Medical School Spine/Paraspinal Sarcoma: Set-Up Verification Set-up Radiograph with Digitized Set-up Points, Actual (red) and Desired (blue) Isocenter Axes Sacral Chondrosarcoma

15. Harvard Medical School Spine/Paraspinal Sarcoma: Treatment Planning Normal tissue constraintsNormal tissue constraints Spinal cordSpinal cord –Surface: 63 CGE Center: 54 CGE (for length up to 5 cm) Cauda equinaCauda equina –Limit dose to contralateral nerves when possible RectumRectum –Use omental flap to displace rectum for sacral chordomas Small bowelSmall bowel –50.4 CGE SkinSkin –Avoid full dose on skin overlying sacrum

16. Harvard Medical School Spine/Paraspinal Sarcoma: Treatment Planning Beam SelectionBeam Selection –Protons not stopped on the spinal cord Concern about end of range uncertaintyConcern about end of range uncertainty RBE variation at end of proton rangeRBE variation at end of proton range –Patch fields contour dose around spinal cord

17. Harvard Medical School Thoracic Chondrosarcoma Treatment Plan (Protons) 60 year old diabetic female with T9 Chondrosarcoma Radiated after laminectomy 72 CGE (IMRT photons 45 Gy protons 27 CGE)

18. Harvard Medical School 90 Y Plaque Irradiation Plaque on Dura

19. Harvard Medical School Results Plaques Comparative depth doses 192 Ir 90 Y 2 mm 80%27% 4 mm (cord surface) 60% 8% 8 mm (cord center) 50% 0%

20. Harvard Medical School Exposed Gafchromic Film Surface Exposure %Depth Dose Exposure

21. Harvard Medical School Spine and Paraspinal Sarcoma Accrual target50 patientsAccrual target50 patients –Statistics Anticipated local failure with lower dose photons estimated to be 50% with spine chordomas at 5 yearsAnticipated local failure with lower dose photons estimated to be 50% with spine chordomas at 5 years 80% power to detect decrease to 30% local failure rate80% power to detect decrease to 30% local failure rate

22. Harvard Medical School Spine and Paraspinal Sarcoma Accrual 50 Patients 12/97-3/2005Accrual 50 Patients 12/97-3/2005 –Chordoma28Chondrosarcoma14 –Liposarcomas 2Angiosarcoma 1 –Ewing’s Sarcoma 1Giant cell tumor 1 –Malignant schwannoma 1Osteosarcoma 1 –Spindle and round cell 1 Thoracic 12Lumbar 12Sacrum 26Thoracic 12Lumbar 12Sacrum 26 Primary 37Locally recurrent 13Primary 37Locally recurrent 13 Extent of surgeryExtent of surgery –Grossly resected 25 –Subtotally excised 12 –Biopsy only 13 All but 1 patient completed Rx (social reasons)All but 1 patient completed Rx (social reasons) –Dose delivery within 3% of protocol target in all patients

23. Harvard Medical School Spine and Paraspinal Sarcoma ResultsResults –Median follow-up: 27 months –Local Recurrence: 6 at 8-32 months after start of XRT 2 also with distant metastases2 also with distant metastases 2 chordomas, 4 chondrosarcomas2 chordomas, 4 chondrosarcomas 3/13 patients treated for recurrent tumor vs. 3/27 for primary tumor3/13 patients treated for recurrent tumor vs. 3/27 for primary tumor –Distant Metastases: 3 isolated at 11-35 m after start of XRT –LR, DM, and tumor-associated deaths were all in pts with gross residual disease except 1 pt with dedifferentiated chondrosarcoma with tumor cut through at initial surgery –Survival 3 died of progressive tumor (2 chondrosarcoma, 1 malig schwannoma)3 died of progressive tumor (2 chondrosarcoma, 1 malig schwannoma) 2 died unrelated causes (cardiac, oral cancer)2 died unrelated causes (cardiac, oral cancer) 1 lost to follow-up1 lost to follow-up

24. Harvard Medical School Sacral Chordoma T1 post-gadolinium sagittal MRI Pre-treatment S3-4 chordoma 77.4 CGE (photons 30.6 Gy protons 46.8 CGE)

25. Harvard Medical School Sacral Chordoma Treatment Plan S3-4 chordoma 77.4 CGE (photons 30.6 Gy protons 46.8 CGE)

26. Harvard Medical School Sacral Chordoma Treatment Plan S3-4 chordoma 77.4 CGE (photons 30.6 Gy protons 46.8 CGE)

27. Harvard Medical School Sacral Chordoma T1 post-gadolinium sagittal MRI S3-4 chordoma 77.4 CGE (photons 30.6 Gy protons 46.8 CGE) No evidence of progressive disease at 19 months

28. Harvard Medical School Unresectable Sacral Osteosarcoma: Axial CT scan 19 year old with S1 Osteosarcoma After 12 weeks of chemotherapy Concurrent chemoradiation starting week 16 70.2 CGE (photons 18 Gy protons 52.2 CGE)

29. Harvard Medical School Unresectable Sacral Osteosarcoma: Treatment Plan (Axial) 19 year old with S1 Osteosarcoma After 12 weeks of chemotherapy Concurrent chemoradiation starting week 16 70.2 CGE (photons 18 Gy protons 52.2 CGE)

30. Harvard Medical School Unresectable Sacral Osteosarcoma: Axial CT scan 19 year old with S1 Osteosarcoma 29 months after treatment Chemotherapy + 70.2 CGE (photons 18 Gy protons 52.2 CGE) No evidence of progressive disease

31. Harvard Medical School Multiply (x4) recurrent G2 chondrosarcoma T4

32. Harvard Medical School Multiply recurrent G2 chondrosarcoma T4

33. Harvard Medical School Multiply recurrent G2 chondrosarcoma T4

34. Harvard Medical School Multiply recurrent G2chondrosarcoma T4 Local recurrence 40 months after Surgery, plaque, XRT

35. Harvard Medical School Multiply (x6) recurrent G2 chondrosarcoma T spine

36. Harvard Medical School Dedifferentiated chondrosarcoma L spine

37. Harvard Medical School Multiply recurrent chondrosarcoma T spine

38. Harvard Medical School Multiply recurrent chondrosarcoma T spine

39. Harvard Medical School Spine and Paraspinal Sarcoma Acute ComplicationsAcute Complications –> Grade 3 1 painful sacral stress fracture without late sequelae.1 painful sacral stress fracture without late sequelae. Late ComplicationsLate Complications –> Grade 3 1 sacral neuropathy: Large, unresected sacral chordoma ( 77.4 CGE) LE weakness, stress urinary incontinence, poor rectal tone at 5.5 years1 sacral neuropathy: Large, unresected sacral chordoma ( 77.4 CGE) LE weakness, stress urinary incontinence, poor rectal tone at 5.5 years 1 erectile dysfunction: 65 y.o. unresected sacral chordoma (77.4 CGE) 4 yrs1 erectile dysfunction: 65 y.o. unresected sacral chordoma (77.4 CGE) 4 yrs –Grade 2 1 erectile dysfunction: 46 y.o. recurrent T3-5 chondrosarcoma and cord compressions (77.4 CGE) at 2 years; responsive to sildenafil1 erectile dysfunction: 46 y.o. recurrent T3-5 chondrosarcoma and cord compressions (77.4 CGE) at 2 years; responsive to sildenafil One late sacral stress fracture after fall, 3 months after XRT and subtotal intralesional excision/MMA packing; managed with nail; chronic painOne late sacral stress fracture after fall, 3 months after XRT and subtotal intralesional excision/MMA packing; managed with nail; chronic pain

40. Harvard Medical School Spine and Paraspinal Sarcoma ConclusionsConclusions –High dose photon/proton XRT can be delivered –Morbidity to date appears acceptable. –Encouraged to date by the treatment results with these challenging tumors Await further follow-upAwait further follow-up Prefer to radiate at time of initial presentationPrefer to radiate at time of initial presentation –Potential concern about late sacral nerve toxicity in patients receiving 77.4 Gy