A Prospective Cardiac Screening Study in Asymptomatic Long-term Survivors of Hodgkin's Lymphoma...
2
Materials/Methods: For 187 HL survivors who developed NSCLC (HL-LC group) and 178,431 patients with a first or only NSCLC (LC-1 group), actuarial OS were compared, accounting for gender (34% of HL-LC group are female), race, sociodemo- graphic status, radiation for HL, calendar year of NSCLC diagnosis, age at NSCLC diagnosis, NSCLC stage and NSCLC histology. All patients were derived from the population-based Surveillance, Epidemiology, and End Results program. Results: The NSCLC stage distribution of the HL-LC and LC-1 groups are similar: ?20% localized, ?30% regional and ?50% distant. For the HL-LC group, with more advanced NSCLC stage, there was a significant trend towards younger age at the time of HL diagnosis (median age 58, 51 and 42 years for localized, regional and distant NSCLC respectively, p \ 0.0001), younger age at the time of NSCLC diagnosis (median age 64, 60 and 58 years, p = 0.014), and longer latency between HL and NSCLC (me- dian 6.9, 10.8 and 14.4 years, p = 0.0003). The HL-LC group was significantly younger than the LC-1 group for each NSCLC stage. For localized and regional NSCLC, the HL-LC group had a greater percentage of patients with squamous cell histology. The OS of each HL-LC stage group was inferior to that of the corresponding LC-1 group. For regional and distant NSCLC, .80% of the deaths among HL-LC and LC-1 patients were from NSCLC; for localized NSCLC, 48% and 64% of deaths among HL-LC and LC-1 patients respectively were from NSCLC (difference not significant). Cox regression analyses of all HL-LC and LC-1 patients demonstrated a significantly worse OS of HL-LC versus LC-1 patients with HRs for death of 1.60 (95% CI 1.08- 2.37, p \ 0.0001), 1.67 (95% CI 1.26-2.22, p = 0.0004), 1.31 (95% CI 1.06-1.61, p = 0.013) for localized, regional and distant NSCLC respectively. Among HL-LC patients with localized NSCLC, only older age at NSCLC diagnosis was a significantly adverse predictor of OS (HR = 1.05, p = 0.035). For HL-LC patients with regional NSCLC, prior radiation for HL (HR = 1.94, p = 0.064), white race (HR = 4.2, p = 0.018) and earlier year of NSCLC diagnosis (HR = 1.08, p = 0.039) were adverse predictors of OS. Conclusions: NSCLC developing after HL is characterized by a poorer OS compared to NSCLC patients with no history of HL. NSCLC stage is the most significant predictor of OS among HL survivors. Given the mortality as well as morbidity of NSCLC, lifelong follow-up of HL survivors is needed, with the understanding that these patients are not only at increased risk of NSCLC, but appear to fare worse with respect to OS. Author Disclosure: M.T. Milano, None; H. Li, None; L.S. Constine, None; L.B. Travis, None. 2763 Limited Stage Mantle Cell Lymphoma: Treatment Outcomes at the Princess Margaret Hospital R. W. Tsang 1 , 2 , M. Bernard 1,2 , L. W. Le 1,2 , D. C. Hodgson 1,2 , A. Sun 1,2 , W. Wells 1,2 , M. Crump 1,2 , M. K. Gospodarowicz 1,2 1 Princess Margaret Hospital, Toronto, ON, Canada, 2 University of Toronto, Toronto, ON, Canada Purpose/Objective(s): Mantle cell lymphoma (MCL) is a rare cancer, with the majority of patients presenting in stage III-IV, and the outcomes are poor (median OS 36-48 months). To determine the curability of localized MCL, we examined treatment outcomes in stage I-II disease at our institution. Materials/Methods: Between 1990-2007, 26 patients (pts) with stage I (38%) and stage II (62%) MCL with a median age of 63 years and M:F ratio of 18:8 were treated. Sites involved were head and neck in 19 (73%), mediastinum - 2 (8%) and pelvis - 5 (19%). Eleven pts had a stage modified IPI $ 2. Five had a blastoid variant of MCL. Five pts were treated with palliative intent. Twenty-one pts (81%) were treated with radical intent: 17 CT+RT, 2 RT, 2 CT followed by autologous stem cell transplant (ASCT), although 1 did not receive ASCT because of lack of response to CT. 13 pts received CHOP, 5 - RCHOP, 1 CVP; and of these, most received $ 6 cycles (79%). The RT median dose was 35 Gy (range, 33.25-40.2 Gy) and the majority received IFRT (15/17). Analysis was focused on 21 patients treated with a curative intent. Results: For 21 pts treated with curative intent, median follow-up was 60.2 months (range, 15-232 months). The overall response rate was 96% (CR: 18 pts, CRu: 1 pt, PR: 1 pt, NR: 1 pt). Among the 19 CR/CRu pts, 9 relapsed for a 5-year relapse rate of 46% (95% CI: 21%-71%). 5-year OS was 62% (95% CI: 43%-89%). Local control was 95%. Median PFS and OS were 3.2 yrs and 6.4 yrs, respectively. Relapses were observed at distant sites (8 pts), in 3 pts with GI involvement. One pt had both local and distant relapse. Following initial treatment failure (relapse/persistent disease, n = 11), the subsequent treatment intent was radical in 2 (ASCT for 1 pt and gastric RT for the other) and palliative in 9. In univariate analysis, blastoid variant and clinical stages II were prognostic factors for PFS, HR 5.9 (95% CI: 1.6-21.5, p = 0.007) and 5.1 (95% CI: 1.1-24.2, p = 0.04), respectively. Mul- tivariate analysis could not be performed due to small sample size. Conclusions: With an aggressive treatment approach for stage I-II MCL using combined CT+RT; local control was achieved in 95%. Systemic relapse remained a significant problem. Stage I-II MCL is associated with better outcomes than reports indicate for stage III-IV pts. Radiation therapy should remain part of curative treatment plan in stage I-II MCL. Author Disclosure: R.W. Tsang, None; M. Bernard, None; L.W. Le, None; D.C. Hodgson, None; A. Sun, None; W. Wells, None; M. Crump, None; M.K. Gospodarowicz, None. 2764 A Prospective Cardiac Screening Study in Asymptomatic Long-term Survivors of Hodgkin’s Lymphoma Treated with Mediastinal Radiation Therapy M. H. Chen 1 , 2 , A. K. Ng 1,2 , T. F. Chu 1 , J. Zhou 1 , K. Gauvreau 1 , P. M. Mauch 2 1 Children’s Hospital Boston, Boston, MA, 2 Brigham and Women’s Hospital/Dana-Farber Cancer Institute, Boston, MA Purpose/Objective(s): To prospectively determine the prevalence of modifiable cardiac risk factors (CRF) and cardiovascular disease (CVD) in Hodgkin’s Lymphoma (HL) survivors treated with mediastinal radiation therapy (RT), and to analyze their association. Materials/Methods: We enrolled 210 consecutive HL survivors (. 5 years since RT completion, and age $ 15 at time of HL diagnosis), treated with RT at the Brigham and Women’s Hospital/Dana-Farber Cancer Institute, and without active symptoms of CVD. Median dose to the mediastinum was 39.6 Gy (range, 25.5 - 53 Gy). Subjects underwent cardiac evaluation including physical examination, resting echocardiography (echo) and stress echo, and lipid profile testing. Modifiable CRFs were defined as hypertension (HTN), hyperlipidemia, diabetes, obesity, smoking within 5 yrs, and physical inactivity. Significant CVD was Proceedings of the 52nd Annual ASTRO Meeting S545
Transcript of A Prospective Cardiac Screening Study in Asymptomatic Long-term Survivors of Hodgkin's Lymphoma...
Proceedings of the 52nd Annual ASTRO Meeting S545
Materials/Methods: For 187 HL survivors who developed NSCLC (HL-LC group) and 178,431 patients with a first or onlyNSCLC (LC-1 group), actuarial OS were compared, accounting for gender (34% of HL-LC group are female), race, sociodemo-graphic status, radiation for HL, calendar year of NSCLC diagnosis, age at NSCLC diagnosis, NSCLC stage and NSCLC histology.All patients were derived from the population-based Surveillance, Epidemiology, and End Results program.
Results: The NSCLC stage distribution of the HL-LC and LC-1 groups are similar: ?20% localized, ?30% regional and ?50%distant. For the HL-LC group, with more advanced NSCLC stage, there was a significant trend towards younger age at the time ofHL diagnosis (median age 58, 51 and 42 years for localized, regional and distant NSCLC respectively, p \0.0001), younger ageat the time of NSCLC diagnosis (median age 64, 60 and 58 years, p = 0.014), and longer latency between HL and NSCLC (me-dian 6.9, 10.8 and 14.4 years, p = 0.0003). The HL-LC group was significantly younger than the LC-1 group for each NSCLCstage. For localized and regional NSCLC, the HL-LC group had a greater percentage of patients with squamous cell histology.The OS of each HL-LC stage group was inferior to that of the corresponding LC-1 group. For regional and distant NSCLC,.80% of the deaths among HL-LC and LC-1 patients were from NSCLC; for localized NSCLC, 48% and 64% of deaths amongHL-LC and LC-1 patients respectively were from NSCLC (difference not significant). Cox regression analyses of all HL-LC andLC-1 patients demonstrated a significantly worse OS of HL-LC versus LC-1 patients with HRs for death of 1.60 (95% CI 1.08-2.37, p \ 0.0001), 1.67 (95% CI 1.26-2.22, p = 0.0004), 1.31 (95% CI 1.06-1.61, p = 0.013) for localized, regional and distantNSCLC respectively. Among HL-LC patients with localized NSCLC, only older age at NSCLC diagnosis was a significantlyadverse predictor of OS (HR = 1.05, p = 0.035). For HL-LC patients with regional NSCLC, prior radiation for HL (HR =1.94, p = 0.064), white race (HR = 4.2, p = 0.018) and earlier year of NSCLC diagnosis (HR = 1.08, p = 0.039) were adversepredictors of OS.
Conclusions: NSCLC developing after HL is characterized by a poorer OS compared to NSCLC patients with no history of HL.NSCLC stage is the most significant predictor of OS among HL survivors. Given the mortality as well as morbidity of NSCLC,lifelong follow-up of HL survivors is needed, with the understanding that these patients are not only at increased risk of NSCLC,but appear to fare worse with respect to OS.
Author Disclosure: M.T. Milano, None; H. Li, None; L.S. Constine, None; L.B. Travis, None.
2763 Limited Stage Mantle Cell Lymphoma: Treatment Outcomes at the Princess Margaret Hospital
R. W. Tsang1,2, M. Bernard1,2, L. W. Le1,2, D. C. Hodgson1,2, A. Sun1,2, W. Wells1,2, M. Crump1,2, M. K. Gospodarowicz1,2
1Princess Margaret Hospital, Toronto, ON, Canada, 2University of Toronto, Toronto, ON, Canada
Purpose/Objective(s): Mantle cell lymphoma (MCL) is a rare cancer, with the majority of patients presenting in stage III-IV, andthe outcomes are poor (median OS 36-48 months). To determine the curability of localized MCL, we examined treatment outcomesin stage I-II disease at our institution.
Materials/Methods: Between 1990-2007, 26 patients (pts) with stage I (38%) and stage II (62%) MCL with a median age of 63years and M:F ratio of 18:8 were treated. Sites involved were head and neck in 19 (73%), mediastinum - 2 (8%) and pelvis - 5(19%). Eleven pts had a stage modified IPI $ 2. Five had a blastoid variant of MCL. Five pts were treated with palliative intent.Twenty-one pts (81%) were treated with radical intent: 17 CT+RT, 2 RT, 2 CT followed by autologous stem cell transplant(ASCT), although 1 did not receive ASCT because of lack of response to CT. 13 pts received CHOP, 5 - RCHOP, 1 CVP; andof these, most received $ 6 cycles (79%). The RT median dose was 35 Gy (range, 33.25-40.2 Gy) and the majority receivedIFRT (15/17). Analysis was focused on 21 patients treated with a curative intent.
Results: For 21 pts treated with curative intent, median follow-up was 60.2 months (range, 15-232 months). The overall responserate was 96% (CR: 18 pts, CRu: 1 pt, PR: 1 pt, NR: 1 pt). Among the 19 CR/CRu pts, 9 relapsed for a 5-year relapse rate of 46%(95% CI: 21%-71%). 5-year OS was 62% (95% CI: 43%-89%). Local control was 95%. Median PFS and OS were 3.2 yrs and 6.4yrs, respectively. Relapses were observed at distant sites (8 pts), in 3 pts with GI involvement. One pt had both local and distantrelapse. Following initial treatment failure (relapse/persistent disease, n = 11), the subsequent treatment intent was radical in 2(ASCT for 1 pt and gastric RT for the other) and palliative in 9. In univariate analysis, blastoid variant and clinical stages IIwere prognostic factors for PFS, HR 5.9 (95% CI: 1.6-21.5, p = 0.007) and 5.1 (95% CI: 1.1-24.2, p = 0.04), respectively. Mul-tivariate analysis could not be performed due to small sample size.
Conclusions: With an aggressive treatment approach for stage I-II MCL using combined CT+RT; local control was achieved in95%. Systemic relapse remained a significant problem. Stage I-II MCL is associated with better outcomes than reports indicate forstage III-IV pts. Radiation therapy should remain part of curative treatment plan in stage I-II MCL.
Author Disclosure: R.W. Tsang, None; M. Bernard, None; L.W. Le, None; D.C. Hodgson, None; A. Sun, None; W. Wells, None;M. Crump, None; M.K. Gospodarowicz, None.
2764 A Prospective Cardiac Screening Study in Asymptomatic Long-term Survivors of Hodgkin’s Lymphoma
Treated with Mediastinal Radiation TherapyM. H. Chen1,2, A. K. Ng1,2, T. F. Chu1, J. Zhou1, K. Gauvreau1, P. M. Mauch2
1Children’s Hospital Boston, Boston, MA, 2Brigham and Women’s Hospital/Dana-Farber Cancer Institute, Boston, MA
Purpose/Objective(s): To prospectively determine the prevalence of modifiable cardiac risk factors (CRF) and cardiovascular disease(CVD) in Hodgkin’s Lymphoma (HL) survivors treated with mediastinal radiation therapy (RT), and to analyze their association.
Materials/Methods: We enrolled 210 consecutive HL survivors (. 5 years since RT completion, and age $ 15 at time of HLdiagnosis), treated with RT at the Brigham and Women’s Hospital/Dana-Farber Cancer Institute, and without active symptomsof CVD. Median dose to the mediastinum was 39.6 Gy (range, 25.5 - 53 Gy). Subjects underwent cardiac evaluation includingphysical examination, resting echocardiography (echo) and stress echo, and lipid profile testing. Modifiable CRFs were definedas hypertension (HTN), hyperlipidemia, diabetes, obesity, smoking within 5 yrs, and physical inactivity. Significant CVD was
S546 I. J. Radiation Oncology d Biology d Physics Volume 78, Number 3, Supplement, 2010
defined as ischemia, obstructive coronary disease, at least moderately severe valve disease, cardiac arrhythmias requiring mechan-ical intervention, and cardiomyopathy with LVEF \50%. Continuous variables were summarized by mean ± standard deviation(SD). Categorical variables were reported as frequencies and percentages. Univariate logistic regression was used to estimate oddsratios (OR) and 95% confidence intervals (CI).
Results: A total of 182 HL survivors (59% female, mean age 44.1 ± 8.7 yrs, follow-up time 15.1 ± 7.2 years) were studied.Overall, 81% (148) have at least 1 modifiable CRF: 26% (48) had HTN, 55% (101) patients had hyperlipidemia, 20% (37)were obese and 43% (79) were physically inactive. Only 6% were current smokers. Seven percent (13) of patients (77%male, mean age 48.1 ± 6.1 years, follow-up time 19.1 ± 8.0 years) were found to have significant CVD. HTN (OR 3.6;95% CI 1.6-11.5. p = 0.03) and hyperlipidemia (OR, 4.8; 95% CI 1.04-22.4, p = 0.04) were associated with significantscreen-detected CVD.
Conclusions: This prospective screening study identified modifiable CRF in 81% of long-term HL survivors, with HTN and hy-perlipidemia being significantly associated with risk for screen-detected CVD. Routine screening of modifiable CRF is prudent inlong-term HL survivors treated with RT. Management of CRF, along with modern RT techniques and lower treatment doses, mayresult in reduced CVD risk for long-term HL survivors.
Author Disclosure: M.H. Chen, None; A.K. Ng, None; T.F. Chu, None; J. Zhou, None; K. Gauvreau, None; P.M. Mauch, None.
2765 Predicted Risk of Radiation-Induced Cancers after Involved Field- and Involved-Node Radiotherapy with
or without Intensity Modulation for Early Stage Hodgkin Lymphoma in Female PatientsD. C. Weber1, S. Johanson2, N. Peguret1, L. Cozzi3, D. R. Olsen4
1Geneva University Hospital-Campus Cluses Roseraie, Geneva 14, Switzerland, 2Institute for Cancer Research, Nor. RadiumHosp., Rikshospitalet University, Oslo, Norway, 3Oncology Institute of Southern Switzerland, Bellinzona, Switzerland, 4Nor.Radium Hosp., Rikshospitalet University, Oslo, Oslo, Norway
Purpose/Objective(s): To assess the excess relative risk (ERR) of radiation-induced cancers (RIC) in Hodgkin lymphoma (HL)female patients treated with 3D conformal (3DCRT), intensity modulated (IMRT) or volumetric modulated arc (RA) radiationtherapy.
Materials/Methods: Plans for 10 early stage HL female patients were computed for 3DCRT, IMRT and RA with involved-field(IFRT) and involved-node (INRT) radiation fields. The organ at risks (OAR) Dose-Volume Histograms (DVHs) were computedand inter-compared for IFRT vs. INRT and 3DCRT vs. IMRT/RA, respectively. The excessive relative risk (ERR) for cancer in-duction in breasts, lungs and thyroid was estimated, using both linear and non-linear-models.
Results: The mean estimated ERR for breast, lung and thyroid were significantly (p \ 0.01) lower with INRT than with IFRTplanning, regardless of the radiation technique delivery used, assuming a linear dose-risk relationship. Using the non-linearmodel, mean ERR were significantly (p\0.01) increased with IMRT or RA when compared to 3DCRT planning for the breast,lung and thyroid using an IFRT paradigm. After INRT planning, IMRT or RA increased the risk of RIC for lung and thyroidonly.
Conclusions: Regardless of the technique delivery, these data suggest that INRT when compared to IFRT planning may reduce theERR of breast, lung and thyroid RIC for HL female patients, when risk is predicted using a linear model. A non-linear dose-riskrelationship suggests that IMRT or RA increase the risk of RIC for these OARs.
Author Disclosure: D.C. Weber, None; S. Johanson, None; N. Peguret, None; L. Cozzi, Varian Medical Systems, F. Consultant/Advisory Board; D.R. Olsen, None.
2766 Mid-treatment Metabolic Tumor Volume Predicts Progression and Death among Patients with
Hodgkin’s DiseaseD. Tseng1, L. P. Rachakonda1, Z. Su1, R. Advani1, S. Horning3, S. A. Rosenberg1, R. T. Hoppe1, A. Quon1, E. E. Graves1,
B. W. Loo1, P. T. Tran2
1Stanford University School of Medicine, Stanford, CA, 2Johns Hopkins University School of Medicine, Baltimore, MD,3Genentech, Inc., San Francisco, CA
Purpose/Objective(s): Considerable evidence has demonstrated the predictive value of positron emission tomography (PET)imaging using fluorodeoxyglucose in patients with lymphomas. We hypothesized that pre- and mid-treatment quantitative PETmetrics may have increased predictive strength as compared to traditional clinical factors such as the International Prognostic Score(IPS).
Materials/Methods: Thirty pediatric and adult patients with Hodgkin’s disease (HD) treated at presentation or relapse had initialstaging and mid-treatment PET-CT scans. The majority of patients (53%) had stage III-IV disease and 67% had an IPS of 2 orgreater. Mid-treatment scans were performed at a median of 55 days from the staging PET-CT, corresponding to 2 cycles of Stan-ford V chemotherapy. Hypermetabolic tumor regions were segmented semiautomatically on the pre-treatment scans using customsoftware and the metabolic tumor volume (MTV), mean standardized uptake value (SUVmean), maximum SUV (SUVmax) andintegrated SUV (iSUV) were recorded. Mid-treatment scans were then registered to the corresponding pre-treatment scans and theidentical regions interrogated. We analyzed whether IPS, PET parameters or the calculated ratio of mid- to pre-treatment PET pa-rameters (e.g., mid-treatment MTV/pre-treatment MTV) were associated with differences in progression free survival (PFS) oroverall survival (OS).
Results: At the time of this analysis, median follow-up of the study group was 50 months. Six of the 30 patients progressedclinically (defined as death, biopsy-proven recurrence or radiographic findings warranting a change in management). SUVparameters from pre-treatment scans were not significant. Two PET parameters from mid-treatment scans were significantlyassociated with OS as determined by Cox proportional hazards: MTV (p = 0.03) and SUVmax (0.008). Two calculated PET
