RTOG/ASCO Presenters Booklet [PDF]

Reference	Study Number
Cooper, J., Pajak, T., Forastiere, A., Jacobs, J., Saxman, B., Kish, J., Kim, H., Cmelak, A., Rotman, M., Machtay, M., Ensley, J., Chao, C., Schultz, C., Lee, N. and Fu, K.: Postoperative Concurrent Radiochemotherapy in High-Risk SCCA of the Head & Neck: Initial Report of RTOG 9501/Intergroup Phase III Trial. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres)	95-01
Seiferheld, W., Mehta, M., DelRowe, J., Macdonald, D., Langer, C., Scott, C., Curran, W. and Yung, W.A.: Five years of glioblastoma multiforme(GBM) phase II trials at the Radiation Therapy Oncology Group (RTOG). Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres)	9417, 9513, 9602, 9710, 9806
Werner-Wasik, M., Scott, C., Curran Jr, W. and Byhardt, R.: Correlation between acute esophagitis and late pneumonitis in patients (pts) with locally advanced non-small cell lung cancer (LA-NSCLC) receiving concurrent thoracic radiotherapy (RT) and chemotherapy: A multivariate analysis of the Radiation Therapy Oncology Group (RTOG) database. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press (acc as poster discussion)	9410, 9204, 9015
Konski, A., Berkey, B., Ang, K.K. and Fu, K.: The Effect of Education Level on Outcome of Patients Treated on Radiation Therapy Oncology Group (RTOG) Protocol 90-03. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. ( acc as poster pres)	90-03
Fisher, J., Scott, C., Scarantino, C., Leveque, F.G., White, R., Rotman, M., Hodson, D., Meredith, R., Foote, R., Brachman, D. and Lee, N.: Phase III Quality of Life Study (QOL): Reduction in Hyposalivation Does Not Improve QOL for Head and Neck (H&N) Cancer Patients Post Radiation Therapy (p-RT). RTOG 9709. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres)	97-09
Chakravarti, A., Winter, K., Wu, C.-L., Kaufman, D., Hammond, E., Parliament, M., Tester, W., Hagan, M., Grignon, D., Heney, N., Sandler, H. and Shipley, W.: Expression of the Epidermal Growth Factor Receptor (EGFR) is Associated with Improved Outcome in Muscle-Invading Bladder Cancer. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres)	8802, 8903, 9506, 9706
Langer, C., Hsu, C., Curran, W., Komaki, R., Lee, J., Byhardt, R. and Sause, W.: Elderly Patients (Pts) With Locally Advanced Non-Small Cell Lung Cancer (La-NSCLC) Benefit From Combined Modality Therapy. Secondary Analysis of Radiation Therapy Oncology Group (RTOG) 94-10. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as poster discussion)	94-10
Roach, M., DeSilvio, M., Lawton, C., Uhl, V., Machtay, M., Seider, M., Rotman, M., Jones, C., Asbell, S., Valicenti, R., Han, S., Thomas, C. and Shipley, W.: Neoadjuvant Hormonal Therapy (NHT) with Whole-Pelvic (WP) Radiotherapy (RT) Improves Progression-Free Survival (PFS): RTOG (Radiation Therapy Oncology Group) 9413, a Phase III Randomized Trial. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press (acc as special oral pres)	9413
Lee, J., Kim, E., Khuri, F., Feng, L., Chamberlain, R.M., Williams, B., Fu, K., Cooper, J., Ang, K.K., Lippman, S.M., Goepfert, H. and Hong, W.K.: Longterm Adherence in a Head and Neck Chemoprevention Trial. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, Am Soc Clin Oncol, pg. Abs. In Press. (acc as poster pres)	9115

Cooper, J., Pajak, T., Forastiere, A., Jacobs, J., Saxman, B., Kish, J., Kim, H., Cmelak, A., Rotman, M., Machtay, M., Ensley, J., Chao, C., Schultz, C., Lee, N. and Fu, K.: Postoperative Concurrent Radiochemotherapy in High-Risk SCCA of the Head & Neck: Initial Report of RTOG 9501/Intergroup Phase III Trial. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres) Purpose: Despite resection and postoperative irradiation, high-risk (2 or more involved lymph nodes, extra-capsular disease and/or microscopically involved mucosal margins of resection) squamous cell carcinomas (SCCAs) of the head and neck (H&N) frequently recur in the tumor bed. We sought to learn if the concurrent administration of cisplatin (CDDP) with postoperative radiation therapy (RT) would improve local-regional control. Methods: Between 9/95 and 4/00, 459 patients who had high-risk SCCAs of the H&N were enrolled in a prospectively randomized phase III trial. Following resection of all detectable disease, 231 patients were randomly assigned to RT (60 - 66 Gy /30 - 33 fx/ 6 - 6.6 weeks) and 228 patients were randomly assigned to identical RT plus CDDP (100 mg/m2 i.v. on days 1, 22 & 43). The primary outcome endpoint was local-regional control. Results: At a median follow-up of 26.6 months (range: 2.5-62.2), the 2-year local-regional control rate is 73.8% for those assigned to RT and 79.2% for the RT plus CDDP group (p=0.16). Two-year overall survival was 56.6 % and 62.9% (p=0.51) and disease-free survival was 42.5% and 54.2% respectively (p=0.049). The incidence of grade 3+ (1) acute toxicity was 32.8% for RT and 74.9% for RT plus CDDP (p < 0.0001, mostly from hematologic, mucous membrane and/or gastrointestinal toxicity from chemotherapy), (2) late toxicity was 15.3% for RT and 19.2% for RT plus CDDP (p = 0.16), and (3) total worst toxicity was 43.6% for RT and 76.4% for RT plus CDDP (p < 0.0001). Conclusion: In the post-operative adjuvant setting, this preliminary analysis demonstrated no significant improvement of local-regional control or overall survival of these high-risk patients from the addition of concurrent cisplatin to radiotherapy. However, disease-free survival was significantly improved, at the cost of significantly increased acute and total toxicity.

Seiferheld, W., Mehta, M., DelRowe, J., Macdonald, D., Langer, C., Scott, C., Curran, W. and Yung, W.A.: Five years of glioblastoma multiforme(GBM) phase II trials at the Radiation Therapy Oncology Group (RTOG). Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres) From 1995 to 2000, the RTOG completed five phase II trials of pharmacological agents combined with standard radiation therapy(RT) for GBM. The RTOG's development of recursive partitioning analysis (RPA) classification for GBM patients enabled a meaningful comparison of single arm trials against the RTOG GBM database of patients treated with BCNU/RT regimens. This approach facilitated the exploration of several agents via phase II trials, rather than focusing resources on a single-agent phase III trial. During this period, the RTOG enrolled 410 eligible patients onto five trials, replacing BCNU with either tirapzamine, topotecan, taxol, beta interferon, or thalidomide. Planned sample size ranged from 53 to 84 patients, as the statistical design was modified over time. None of the experimental agents from these studies demonstrated statistically significant improvement in survival from the historical control after adjusting for RPA class. On the other hand, it is important to realize that several of the studies exhibited survival similar to the historical control, and with fewer life-threatening toxicities. These results question the standard practice of concurrent BCNU for GBM patients. Furthermore, with the observed low toxicity levels, phase II testing of doublet or triplet combinations of these and other agents with RT is feasible, and may yield better results than single-agent /RT approaches.   Median Survival (mo.)   Overall RPA III-IV RPA V-VI Study N >G4 Tox. Obs. Exp. Obs. Exp. Obs. Exp. 9417:Tirapazamine Arm 1: 159mg/m2 53 2% 10.8 10.8 12.8 12.9 7.9 8.0 9417:Tirapazamine Arm 2: 260mg/m2 68 11% 9.5 9.7 11.5 12.0 5.0 8.1 9513: Topotecan 79 54% 9.3 10.1 10.3 12.0 5.5 8.2 9602: Paclitaxel 61 13% 9.7 9.6 11.1 12.6 6.4 7.7 9710: ß-Interferon 60 2% 13.2 10.3 14.1 12.6 10.7 8.2 9806: Thalidomide 89 8% 10.0 9.5 14.4 12.4 4.3 7.9 Obs. = Observed, Exp. = Expected (from historical control)

Werner-Wasik, M., Scott, C., Curran Jr, W. and Byhardt, R.: Correlation between acute esophagitis and late pneumonitis in patients (pts) with locally advanced non-small cell lung cancer (LA-NSCLC) receiving concurrent thoracic radiotherapy (RT) and chemotherapy: A multivariate analysis of the Radiation Therapy Oncology Group (RTOG) database. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press (acc as poster discussion) Purpose: To investigate the time course, predisposing factors and correlation between the acute esophagitis (ES) and late pneumonitis (PN) in pts with LA-NSCLC receiving concurrent RT and chemotherapy. Method: RTOG database was analyzed using logistic regression to identify factors associated with severe (Grade ³3) ES or severe (Grade ³3) PN. Additionally, polychotomous regression was used to provide odds ratios for the onset of each grade of ES or PN. Results: A total of 585 patients from 4 RTOG trials were analyzed. RT was either standard fractionated (60 Gy) or hyperfractionated (69.6 Gy). Chemotherapy was cisplatin-based. Grade ³2 ES developed in 76% of pts and Grade ³3 ES, in 37%. ES grade peaked in the first month in 38% of pts, in the second month in 49%, and in the third month, in 3%. PN peaked by 6 months in 44% of pts, between 6 and 12 months in 14%, and between 12 and 18 months, in 4%. Sixty six percent of all pts developed ³Grade 2 and 19%, ³Grade 3 PN by 4 years. In regression analysis, poor pulmonary function, tumor dose, dose to the esophagus, smaller primary tumor and absence of cough were significantly associated with severe PN. No factors were predictive of severe ES. The following factors were associated with any ES: higher tumor dose and the subsequent development of severe PN (p=0.00008 and 0.018, respectively). Conclusion: Acute ES develops within 2 months in practically all pts, whereas late PN can occur up to 18 months and beyond. The actuarial incidence of PN in long surviving LA-NSCLC pts receiving concurrent chemoradiotherapy is high. Pts with severe ES are not more likely than the remainder of pts to develop late PN. However, severe late PN seems more likely to be associated with the preceding severe acute ES suggesting an individual predisposition. (Medimmune, Inc. and RTOG U10 CA21661, CCOP U10 CA37422, Stat U10 CA32115).

Konski, A., Berkey, B., Ang, K.K. and Fu, K.: The Effect of Education Level on Outcome of Patients Treated on Radiation Therapy Oncology Group (RTOG) Protocol 90-03. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. ( acc as poster pres) Purpose: It has been hypothesized people in lower socioeconomic groups have worse outcomes because they present with more advanced stage cancers or receive inferior treatment. We investigate this hypothesis by using education level as a proxy for income level in patients treated on RTOG protocol 90-03. Methods: RTOG 90-03 was a phase III randomized trial investigating 4 different radiation fractionation schedules in the treatment of locally advanced head and neck cancers. Overall survival and local-regional control rates were analyzed by education level as measured by patient response on the demographic form at study entry. Education level is separated into the following categories: no answer, no school, grade 1-8, some high school, high school graduate/GED, and college/technical. Results: No significant difference existed in the distribution of patients by education level when analyzed by tumor stage, nodal status or primary site. A significant difference was noted in the distribution of patients by education level between the standard fractionated radiation treatment arm and the hyperfractionated radiation treatment arm (p=0.018). No statistical difference was noted in median treatment days or median radiation dose in all four treatment groups when separated by education level. Patients attending college had highly significantly better overall survival and local-regional control outcome than the other groups combined (p=0.0030 and p=0.0052 respectively: from Cox proportional hazards model with educational level, assigned treatment, t-stage, n-stage, KPS, and primary site). Conclusion: Patients attending college or technical school had improved overall survival and local-regional control than those who did not. These differences cannot be explained by differences in tumor stage or treatment. Other possible factors such as poorer overall health or lack of support systems contributing to these results need to be further investigated.

Fisher, J., Scott, C., Scarantino, C., Leveque, F.G., White, R., Rotman, M., Hodson, D., Meredith, R., Foote, R., Brachman, D. and Lee, N.: Phase III Quality of Life Study (QOL): Reduction in Hyposalivation Does Not Improve QOL for Head and Neck (H&N) Cancer Patients Post Radiation Therapy (p-RT). RTOG 9709. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres) Purpose: To determine if the prevention of hyposalivation following curative radiation therapy (RT) for H&N cancer improves patients QOL. Method: Patients were required to receive at least 50 Gy to 50% of the volume of the major salivary glands; provide unstimulated and stimulated saliva samples and complete the Univ. of Washington H&N QOL tool (HNSS) prior to RT, and 3 mo., and 6 mo. p-RT. Patients were randomized to receive, Pilocarpine (P), 5mg qid or placebo. Results: 249 patients were randomized between 3/98 to 1/00, 246 were eligible for analysis for QOL. Patients were evenly distributed between arms by race, gender, tobacco use, tumor site and T-stage (50% T2/T3) and salivary function. KPS 90% was more common in P arm. 20% of patients on P arm and 29% patients on placebo arm were on oral supplements. Use of oral supplements related to activity, employment, eating and mucosal complaints. Placebo arm patients had greater mouth pain and chewing difficulties. Compliance for HNSS was 75% at 3 mo. and 50% at 6mo. There was no difference between treatment arms in any of the HNSS (taste impairment, activity, pain) at 3 mo. or 6mo.despite statistically significant (p=.047 and p=0.049 respectively) preservation of salivary function in the P arm with continued functional loss in the placebo arm. Patients on P arm reported difficulties with swallowing (75%), activity (80%), hyposalivation (64%) and taste (81%). There was no difference between arms at 3 mo. in mucositis scores with both arms demonstrating increased requirement for oral nutrients. Conclusion: The objective prevention of hyposalivation did not affect patients' assessment of salivary function or QOL most likely due to the greater impact mucositis plays in QOL post- rt.

Chakravarti, A., Winter, K., Wu, C.-L., Kaufman, D., Hammond, E., Parliament, M., Tester, W., Hagan, M., Grignon, D., Heney, N., Sandler, H. and Shipley, W.: Expression of the Epidermal Growth Factor Receptor (EGFR) is Associated with Improved Outcome in Muscle-Invading Bladder Cancer. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as oral pres) Introduction: Erb-1 (EGFR) and Erb-2 (Her-2) are two of the best characterized members in the epidermal growth factor receptor pathway. In many tumor types overexpression of these proteins is associated with enhanced malignant potential. Our objective in this study was to investigate the clinical relevance of EGFR and her-2 expression in bladder cancer cases from 4 prospective Radiation Therapy Oncology Group (RTOG) bladder preservation trials using cisplatin-containing chemoradiation (RTOG 8802, 8903, 9506, and 9706). Methods: Tumors from seventy-three cases from patients with muscle-invading T2-T4a bladder cancers had slides interpretable for EGFR staining; fifty-five cases had slides interpretable for her-2 staining. Staining and interpretation of staining were done in a blinded manner, without knowledge of clinical outcome. Staining was judged as positive or negative. Subsequently, staining was correlated with clinical outcome. Results: On univariate analysis, EGFR positivity was significantly associated with improved overall survival (p=0.044); disease-free survival (DFS) (p=0.015); DFS with intact bladder (0.0048); disease-specific survival (DSS) (p=0.042); and DSS with intact bladder (p=0.021). There was also a borderline association between EGFR expression and reduced frequency of distant metastasis (p=0.06). On multivariate analysis adding tumor stage, tumor grade, whether a visibly complete TURBT was done or not and patient age to the model, EGFR positivity was significantly associated with improved DFS. On univariate analysis, Her-2 positivity was significantly associated with reduced complete response rates (p=0.026) after chemoradiation which remained significant on multivariate anlaysis. Conclusion: EGFR expression appears to correlate significantly with improved outcome in bladder cancer, whereas Her-2 expression is significantly associated only with reduced complete response rates after chemoradiation. Further investigations are warranted into how EGFR family members regulate response to chemoradiation in bladder cancer and their potential therapeutic implications.

Langer, C., Hsu, C., Curran, W., Komaki, R., Lee, J., Byhardt, R. and Sause, W.: Elderly Patients (Pts) With Locally Advanced Non-Small Cell Lung Cancer (La-NSCLC) Benefit From Combined Modality Therapy. Secondary Analysis of Radiation Therapy Oncology Group (RTOG) 94-10. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press. (acc as poster discussion) Up until 1994, elderly pts on RTOG trials evaluating bi-modality therapy (Tx) in LA-NSCLC did not appear to benefit from increased Tx intensity or combined modality Tx compared to radiation (TRT) alone (Lung Cancer 29 (suppl 1): 104, 2000, A-340). To determine if this observation still holds, we evaluated outcome in elderly vs. younger pts enrolled on RTOG 94-10, a phase III trial comparing concurrent chemotherapy (CHM) and thoracic radiotherapy (TRT) to standard sequential (SEQ) CHM and TRT. Pts on the SEQ arm received cisplatin (P) 100 mg/m2 d 1 and 29, vinblastine 5 mg/m2 Q wk X 5, followed by 60 Gy TRT starting d 50; Arm 2 used identical CHM and 60 Gy TRT starting d 1 (CON-QD). Arm 3 employed P 50 mg/m2 d 1, 8, 29, and 36, and VP-16 50 mg BID d 1-5, 8-12, 29-33, 36-40; and 69.6 Gy (1.2 Gy BID) starting d 1 (CON-BID). From 1994 to 1998, 610 pts were enrolled; 595 (97.5%) were analyzable. 104 (17%) were > 70 yrs vs 488 < 70 yrs. Demographics were similar for each age cohort. Grade > 3 neutropenia and max toxicity/pt were more pronounced in the elderly in each arm. Gr > 3 esophagitis in pts receiving CON-QD and CON-BID was higher in the elderly (33% and 60%, respectively) vs. those <70 yrs (23% and 42%, respectively). Median survival (MST) in the elderly favored concurrent chemoradiation: 22.4 mos for CON-QD vs. 16.4 mos for CON-BID vs. 10.8 mos for SEQ (p=0.069), while MSTs for those <70 were 15.5, 16, 15.7 mos, respectively. Tx completion rate was significantly lower in the elderly vs. younger pts receiving CON-BID: 69% vs. 87% (p<0.001); no differences in Tx completion rates were observed for SEQ or CON-QD. Long-term toxicities were similar between those <70 and > 70 yrs. Conclusions: In contrast to historic precedent, fit, elderly pts enrolled on RTOG 9410 benefited from concurrent chemoradiation. CON-QD therapy appeared superior to CON-BID therapy with respect to survival and esophagitis rates. Short-term toxicities were acceptable, though more pronounced in the elderly.

Roach, M., DeSilvio, M., Lawton, C., Uhl, V., Machtay, M., Seider, M., Rotman, M., Jones, C., Asbell, S., Valicenti, R., Han, S., Thomas, C. and Shipley, W.: Neoadjuvant Hormonal Therapy (NHT) with Whole-Pelvic (WP) Radiotherapy (RT) Improves Progression-Free Survival (PFS): RTOG (Radiation Therapy Oncology Group) 9413, a Phase III Randomized Trial. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, J Clin Oncol, pg. Abs. In Press (acc as special oral pres) Purpose: Compare WP RT to Prostate only (PO) RT and NHT to Adjuvant hormonal therapy (AHT) in men with high-risk localized prostate cancer. Materials and Methods: From April 1, 1995 until June 1, 1999 1323 men were randomized to four arms comparing WP RT+NHT, PO RT+NHT, WP RT+AHT and PO RT+AHT. Eligibility included localized adenocarcinoma of the prostate with an elevated prostate specific antigen (PSA) <100 ng/ml. and an estimated risk of lymph node (LN) involvement >15%, (LN = (2/3) PSA +[(GS-6) x 10]). Hormonal therapy consisted of flutamide 250 mg p.o. TID and goserelin acetate or leuprolide depot administered two months before and during RT (NHT) or for 4 months following the completion of RT (AHT). WP RT consisted of 50Gy (minimum field size 16 x 16 cm); PO RT required a maximum field size < 11 x 11 cm, with 70 Gy delivered to all arms. PFS was defined as death (any cause), local progression, or regional, distant, or PSA failure. Results: With a median follow-up of 59.3 months patients treated with NHT and WP RT experienced a 4 year PFS of 59.6% compared to 44.3% when treated with PO RT (p=0.001). However, patients treated with AHT experienced a similar progression free survival with WP or PO RT (p=0.87). Longer follow-up will be required to address the impact of WP+NHT on overall survival. Conclusions: Prophylactic WP RT and NHT reduced the risk of disease progression for patients with high-risk prostate cancer. However, NHT combined with RT was no more effective than AHT when only the prostate was irradiated and WP RT was no more effective than PO RT without NHT. Future studies attempting to identify other active agents should also consider the possibility of interactions in pelvic lymph nodes and the impact of the volume treated and sequence of drug administration on progression free survival. Supported by grants (RTOG U10 CA21661, CCOP U10 CA37422, Stat U10 CA32115) from the National Cancer Institute. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.

Lee, J., Kim, E., Khuri, F., Feng, L., Chamberlain, R.M., Williams, B., Fu, K., Cooper, J., Ang, K.K., Lippman, S.M., Goepfert, H. and Hong, W.K.: Longterm Adherence in a Head and Neck Chemoprevention Trial. Proceeding from Am Soc Clin Oncol (ASCO), Orlando, FL, Am Soc Clin Oncol, pg. Abs. In Press. (acc as poster pres) The retinoid head and neck second primary trial, an intergroup, placebo-controlled, double-blind chemoprevention trial was launched in 1991. Eligible patients (pts) had previously treated stage I or II head and neck cancers diagnosed within 3 years (yrs) and were treated with low dose 13-cis-retinoic acid. The primary endpoint is development of second primary tumors (SPT). An eight week run-in period was used to improve study adherence, with pts >75% drug adherent eligible for randomization. This trial has completed accrual with 1384 registered pts and 1191 randomized and eligible pts. 10.8% of registered pts were not randomized. 1249 (96%) of 1299 pts evaluable had a run-in adherence rate of 75% or greater (table). Of the 1191 randomized and eligible pts, the percentage of pts remaining on study are: 93%, yr 1; 92%, yr 2; 77%, yr 3; 66%, yr 5; 50%, yr 7. Reasons pts came off study included pt refusal, new primary, non-compliance, death and lost to follow-up. Adherence to study drug during the trial was recorded as pill counts and correlated with patient calendars. Pts were required to take drug for 3 yrs unless they developed recurrence or an SPT. During the 3-yr-treatment period, pill compliance is greater than 93%. Overall dose intensity during the 3-yr-treatment period is greater than 90%. Overall, we have excellent compliance in the study. The run-in phase appears to be effective in selecting those patients who are sufficiently compliant for randomization. Number of Patients in the Trial Status Run-in Rand. Yr 0-1 Yr 1-2 Yr 2-3 Yr 3-5 Yr 6-7 Beginning of period 1338 1193 1191 1104 1014 918 789 Off study 145 2 87 90 96 129 199 End of period 1193 1191 1104 1014 918 789 590 Study Compliance (N*) 1299   990 860 698     % of compliance 75% 96%   93% 94% 93%     *Number of pts with compliance data