Tuscano, Joseph M.

Abstract Abstract 1603 Background: We have previously reported similar outcomes for patients with bulky stage I or II mediastinal HL treated with combined modality therapy either with ABVD + radiotherapy or the Stanford V regimen in the North American Intergroup trial E2496 (Advani et al, ASH 2010 abstract 416). In the current analysis, we compare the patterns of failure between the two groups. Methods: Patients with stage I/II bulky mediastinal HL (maximum mediastinal mass width > 1/3 of intrathoracic diameter) were randomized to receive chemotherapy (CT) on either Arm A (ABVD x 6–8 cycles administered q 28 days) or Arm B (12 weeks of Stanford V administered weekly). Two-3 weeks after completion of chemotherapy all patients received modified involved field radiotherapy (RT) (36 Gy) delivered to the mediastinum, hila, and supraclavicular regions. Patients on Stanford V arm also received involved field RT to any other sites ≥ 5 cm at diagnosis. Patients were assessed 3, 6 and 12 months after completing RT with computed tomography scanning and then every 6 months for 5 years. The primary end points were failure free survival (FFS) and overall survival (OS). Disease progression was defined as 'in-field', 'distant' or both relative to the radiation fields prescribed in the E2496 protocol. Distant sites of failure were further characterized as intra-thoracic, intra-abdominal or other (bone, bone marrow and axillae, if not previously irradiated). Results: Two hundred and sixty-seven patients were randomized: 136 on the ABVD arm and 131 on the Stanford V arm. Patient characteristics were well matched with no differences between two arms in overall response rates (ORR), FFS and OS. (Advani et al ASH 2010 abstract 416). At a median follow up of 5.5 years 40 patients have relapsed with no difference in ABVD (n=18, 13%) versus Stanford V (n=22, 17%) (p=0.49). Central review of RT fields available in 37/40 patients found major violations with under treatment of tumor noted in 7/37 (19%). Patterns of failure are shown in Table 1. There were no differences in patterns of relapse for the two study arms. In-field relapses occurred in <10% in both study arms. Conclusion: For patients with stage I/II bulky mediastinal HL, combined modality therapy with either ABVD +RT or the Stanford V regimen results in excellent disease control. In-field relapse was uncommon. These results set a benchmark for assessing ongoing trials omitting RT in patients with stage I/II bulky mediastinal HL. Future research efforts should focus on risk stratification to identify the small subset of patients who are likely to fail standard upfront therapy. US cooperative group efforts in this subset of patients are ongoing that use interim PET-CT imaging based risk-adapted strategies. Disclosures: Horning: Genentech: Employment, Equity Ownership.

Abstract Abstract 415 Background: The ability to cure patients (pts) with advanced Hodgkin's Lymphoma (HL) with combination chemotherapy (CC) (MOPP and variants, ABVD and variants) represented a major milestone in oncology research, and CC became a paradigm for other malignancies. Further, the rationale for combined modality therapy (CMT) (radiation (RT) and CC) in HL evolved based on the high frequency of relapse in initially involved sites. As response rates and survival improved, newer treatments such as the combined modality Stanford V regimen were developed to shorten the duration of chemotherapy, add RT to sites of disease and reduce toxicity while maintaining or improving the cure rate. Indeed, the Stanford V regimen was tested and validated in a Phase II co-operative group trial (E1492) (J Clin Oncol 2000; 18:972). In order to investigate this approach against “standard” therapy, we conducted a randomized Phase III Intergroup trial of ABVD vs. the Stanford V regimen for patients with locally extensive or advanced HL. Objectives: The trial was designed to detect a 33% reduction in the failure free survival (FFS) hazard rate with Stanford V compared with ABVD, which corresponds to a difference in five-year FFS of 64% vs. 74%. Method: Patients with locally extensive (defined as clinical Ann Arbor Stage I-IIA/B and bulky mediastinal disease (BMD) (mass > 1/3 maximum intrathoracic diameter on standing postero-anterior chest x-ray or >/−10 cm on computerized tomography) or advanced (Ann Arbor Stage III or IV) HL were randomized to receive either ABVD × 6–8 cycles (C) (51% had 6 C, 35% had 8 C, 14% had <6 C) + 36 Gy (only in pts with BMD) or Stanford V × 12 weeks (95% had 12 weeks) + 36 Gy (for sites >5cm or for macroscopic splenic disease). The log-rank test was used to compare FFS for all eligible patients stratified on extent of disease (locally extensive vs. advanced), and number of International Prognostic Factor Project (IPFP) risk factors (0–2 vs. 3–7). An extended Cox model was also used to address non-proportional hazard between the two arms. Results: 854 pts enrolled from April, 1999 to June, 2006 and 812 were eligible for analysis. 404 pts were randomized to ABVD and 408 to Stanford V. Median age was 33 yrs in both arms (range 16–83). 53% were men and 47% women; 4% had Stage I, 31% had Stage II, 39% had Stage III and 25% had Stage IV disease by Ann Arbor criteria. 35% of pts on ABVD and 35% on Stanford V had BMD. Three % of pts had nodular lymphocyte predominant HL, 77% of pts had nodular sclerosis HL, 14% had mixed cell HL. Age, stage, pathology and risk factors (0–2 vs. 3–7) were similar in both arms. In total, 65% were IPFP score 0–2 and 33% were 3–6. Response rate. There was no difference in response rates (RR) between the two arms (ABVD=72% CR+ CCR, 7.7% PR, 7.9% SD; Stanford V= 69 % CR +CCR, 7% PR and 10 % SD. 8% were not evaluable for response on ABVD and 9% on Stanford V. Toxicity was similar in both groups. The most frequent Grade 3 + 4 toxicity was neutropenia, and was similar between the 2 groups (76% Grade 3 + 4 in ABVD and 70% Grade 3+ 4 in Stanford V). Grade 5 toxicity was <1% in both groups. There was, however, more Grade 3 lymphopenia in Stanford V (78% vs. 42%, p< 0.0001) and more Grade 3 +4 sensory neuropathy in Stanford V (10%) than in ABVD (3%) (p< 0.0001). A total of 26 second primary cancers developed, 12 after ABVD and 14 after Stanford V (p=NS). FFS and OS. For 812 eligible patients, with a median follow up of 5.25 years, 5-year FFS was 73% for ABVD and 71% for Stanford V (p=0.29 by log rank) (Figure left); 5-year OS was 88% for ABVD and 87% for Stanford V (p=0.87 by log-rank, HR=0.97, 95% CI: 0.65 to 1.44) (Figure right) indicating no significant difference in either FFS or OS between the 2 arms. Conclusion: In the largest Phase III intergroup trial of HL in North America, there was no significant difference in RR, FFS, OS, and 5-year toxicity when ABVD (+ RT for BMD) is compared with Stanford V (+RT for nodal sites >5 cm and macroscopic splenic disease). There was more Grade 3 lymphopenia (p< 0.0001) and more Grade 3 + 4 sensory neuropathy (p< 0.0001) on Stanford V. Thus ABVD (plus RT for BMD) remains the standard of care because Stanford V did not meet the objective of 33% improvement in FFS. For some patients, Stanford V, when given as described with RT, remains an acceptable alternative. Disclosures: Friedberg: Genentech: Honoraria. Blum:Seattle Genetics: Research Funding; Novartis: Research Funding; Celgene: Research Funding. Horning:Genentech: Employment.

Abstract Abstract 416 Background: Standard therapy for locally extensive Hodgkin Lymphoma (HL) defined as stage I/II with bulky mediastinal disease [mediastinal mass ratio (MMR) > than 1/3 of the chest diameter on standing postero-anterior chest x-ray or 10 cm on computerized tomography] is combined modality therapy (CMT). E2496 was a North American intergroup, randomized phase III study comparing ABVD versus the Stanford V regimen for patients with locally extensive and advanced stage HL. In this subgroup analysis we compare two CMT approaches, ABVD + radiotherapy (RT) and the Stanford V regimen, in patients with stage I/II bulky mediastinal HL. Methods: Patients with stage I/II HL bulky mediastinal disease were randomized to receive chemotherapy (CT) on either Arm A (ABVD × 6–8 cycles administered q 28 days) or Arm B (12 weeks of Stanford V, administered weekly). Two-3 weeks after completion of chemotherapy, modified involved field RT was delivered at 36 Gy to the mediastinum to patients on ABVD as well as Stanford V. Patients on Stanford V also received involved field RT to any other sites >5 cm at diagnosis. The primary endpoint was failure free survival (FFS) defined as the time to either progression/relapse or death. The log-rank test was used to compare FFS and OS (overall survival) for all eligible patients. Results: Of the 812 eligible patients with advanced HL enrolled on the study 267 had locally advanced bulky mediastinal disease: 136 on ABVD and 131 on Stanford V. Patient characteristics between the two randomized groups were well matched with a median age of 30 years and 86% had stage II disease. On ABVD 61% of patients received 6 cycles, 29% 8 cycles of CT and 77% required some dose modification. In Stanford V 97% completed the assigned 12 weeks of CT and 76% required some dose modification. 82% received RT per protocol in ABVD versus 88% in Stanford V. The overall response (CR+PR) was 82% in ABVD and 86% in Stanford V. At a median follow-up of 5.47 years, there are 19 failures and 6 deaths in the ABVD group and 25 failures with 9 deaths in the Stanford V group. We failed to detect statistically significant differences between ABVD +RT and Stanford V for FFS (5y 85% versus 77% p=0.13, HR=1.56 95% CI (0.87, 2.88) and OS (5y 95% versus 92% p=0.31, HR=1.69 95% CI (0.60, 4.75). No difference in hematologic toxicity was observed between the two arms. Evaluation of pulmonary function and patterns of failure are pending. Conclusions: For stage I/II patients with bulky mediastinal disease, CMT with either ABVD +RT or the Stanford V regimen results in high cure rates and is highly effective. ABVD for 6–8 cycles plus 36 Gy RT remains the US standard of care. Longer follow up is required to assess RT related late effects. Future research efforts should focus on risk stratification to identify; a) the 15–20% of patients destined to relapse after standard therapy and evaluate treatment intensification strategies and b) the 80–85% of patients who are cured with standard therapy and determine whether a subset can achieve the same excellent outcomes with a reduction or elimination of radiation. Planned US cooperative group trials will address these questions, using ABVD as the chemotherapy backbone in conjunction with interim PET imaging for risk stratification. Disclosures: Blum: Seattle Genetics: Research Funding; Novartis: Research Funding; Celgene: Research Funding. Horning:Genentech: Employment.

Abstract Abstract 2699 Poster Board II-675 Introduction: High-dose chemotherapy with autologous stem cell transplant (SCT) is a standard treatment option for younger patients with aggressive non-Hodgkin's lymphomas (a-NHL) who fail to respond, or relapse after initial therapy. For patients who relapse after SCT there are few effective treatment options and prognosis remains poor (Vose et al Blood. 80(8):2142–8, 1992). A second SCT was shown to achieve only marginal responses and at an increased risk of toxic death (Lenain et al Hematol J. 5(5):403–9, 2004). Therefore, new approaches are needed for treatment of patients in this poor prognostic group. Lenalidomide (Revlimid®) is an immunomodulatory agent that has shown clinical activity in treatment of several B-cell malignancies. Two phase 2 studies of lenalidomide monotherapy in patients with relapsed or refractory (R/R) a-NHL were conducted in the US (NHL-002; Wiernik et al JCO 26:4952–7, 2008) and internationally (CC-5013 NHL-003). In an extended follow-up of the NHL-002 study, the overall response rate (ORR) was 35%, which included 12% of patients with a complete response (CR), and a median duration of response (DR) lasting 10.4 months (Wiernik et al 2008 EHA. Abst: 764). Comparable efficacy was observed in the larger confirmatory, phase 2, NHL-003 study of lenalidomide in a similar patient population. The goal of this analysis was to learn the ORR and DR to lenalidomide in patients with a prior SCT. Methods: Patient data from both phase II studies were pooled for this report. Eligibility for both studies was similar – patients with R/R a-NHL, which included diffuse large B-cell lymphoma (DLBCL), mantle cell lymphoma (MCL), transformed lymphoma (TL), and follicular lymphoma grade III (FL-III), with measurable disease (≥ 2 cm), and ≥ 1 prior treatment regimen. Patients received oral lenalidomide 25 mg once daily on days 1–21 of a 28-day cycle. Protocol defined treatment continued for up to 52 wks in NHL-002, or until disease progression in NHL-003. Primary endpoint was ORR; secondary endpoints included DR, progression-free survival (PFS), and safety. Results: 87 patients with a prior SCT (14 patients from NHL-002; 73 patients from NHL-003) were included in this analysis. Median age of patients was 61 yrs (range, 23–76), with a median of 4 prior therapies (range, 2–12). The ORR to lenalidomide was 39% (34/87), with 13% (11/87) CR/CRu and 26% (23/87) partial response (PR). Median PFS for all 87 patients was 3.8 months (95% CI, 2.6, 5.6) and the DR for 34 responders was 9.7 months (95% CI, 4.2, 16.3). Responses occurred in 15 of 52 patients (29% ORR; 10% CR/CRu) with DLBCL, 12 of 19 patients with MCL (63% ORR; 26% CR/CRu), and in 6 of 10 patients with TL (60% ORR; 10% CR/CRu). The table summarizes responses for patients who did not have a transplant, for those who had a SCT anytime prior to lenalidomide; as the last therapy prior to lenalidomide; and not as last therapy prior to lenalidomide. Most common grade 3 or 4 adverse events were neutropenia (44%), thrombocytopenia (33%), and anemia (9%). Eighteen (20.6%) patients discontinued treatment due to adverse events. Patients with a prior history of SCT appeared to have a significantly higher rate of thrombocytopenia, all grades (51% v 30%; P = 0.001), and grades 3 or 4 (33% v 16%; P = 0.002). Patients with a prior SCT were also more likely to experience a dose reduction/interruption due to thrombocytopenia compared with those without a prior SCT (25% v 14%; P = 0.027). Conclusions: Based on a pooled dataset from two phase II studies, oral lenalidomide monotherapy appears to be a well tolerated and active therapy resulting in durable responses in patients with R/R a-NHL who had a prior SCT. Furthermore, the potential of achieving a response to lenalidomide appears to be independent of prior history of SCT. Disclosures: Vose: Celgene: Research Funding. Off Label Use: Lenalidomide for the treatment of relapsed/refractory aggressive non-Hodgkin's lymphoma. . Czuczman:Celgene: Research Funding. Zinzani:Celgene: Research Funding. Tuscano:Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees. Pietronigro:Celgene: Employment, Equity Ownership. Ervin-Haynes:Celgene: Employment. Witzig:Celgene: Research Funding.

Abstract Background: Lenalidomide (Revlimid®), an immunomodulatory drug (IMiD®), was recently approved in the US for treatment of relapsed/refractory multiple myeloma and myelodysplastic syndromes associated with a deletion 5q[31] cytogenetic abnormality. Lenalidomide also has demonstrated activity in chronic lymphocytic leukemia and cutaneous T-cell lymphoma while thalidomide, a less potent IMiD®, has activity in non-Hodgkin’s lymphoma as both monotherapy and in combination with rituximab. Aim: To assess the safety and activity of lenalidomide monotherapy in subjects with relapsed/refractory aggressive non-Hodgkin’s lymphoma (NHL). Methods: Subjects with relapsed/refractory aggressive NHL following at least 1 prior treatment regimen with measurable disease are eligible. Subjects receive 25 mg lenalidomide orally once daily on Days 1–21 every 28 days and continue therapy for 52 weeks as tolerated or until disease progression. Response and progression are evaluated using the IWLRC methodology. Results: As of July 25, 2006, 32 subjects of a planned 40 have enrolled and 31 have received drug. Twenty-two subjects are currently evaluable for tumor response. The median age of the 22 response-evaluable subjects is 65 (46–83) and 13 are female. Histology is diffuse large B-cell lymphoma [DLBCL] (n=12), follicular center lymphoma grade 3 [FL] (n=3), mantle cell lymphoma [MCL] (n=5) and transformed [TSF] (n=2). Median time from diagnosis to lenalidomide monotherapy is 2.3 (0.7–7) years and median number of prior treatment regimens per subject is 2 (1–6). Seven subjects (32%) exhibited an objective response (2 complete responses unconfirmed (CRu) and 5 partial responses (PR)), 6 had stable disease (SD) for a tumor control rate (TCR) of 59% and 9 progressive disease (PD). Responses were produced in each of the aggressive histologic subtypes studied: DLBCL (3/12), FL (1/3), MCL (2/5) and TSF (1/2). Five of 11 subjects (45%) with 1–2 prior treatment regimens had an objective response, as did 2 of 3 subjects (67%) who had received a prior stem cell transplant. Median time-to-response was 2 months. Grade 3 or 4 adverse events occurred in 18 of 31 (58%) subjects receiving drug. These were predominantly Grade 3 hematological events (neutropenia, thrombocytopenia) with only 4 subjects (13%) experiencing a Grade 4 adverse reaction. Conclusion: Preliminary results indicate that lenalidomide monotherapy is active with manageable side effects in relapsed/refractory aggressive NHL.

Abstract Background: As a single agent, rituximab produces an overall response rate (ORR) of 20-50% in patients with relapsed/refractory (r/r) B-cell non-Hodgkin lymphomas (NHL). Acquired resistance occurs, however, partially due to defective host immune effector mechanisms. Efforts to augment immune function have included combining rituximab with immune modulators such as IL-2, IL-12, interferonand lenalidomide. These efforts have resulted in improved efficacy of rituximab. Ipilimumab is a human IgG1κ monoclonal antibody (mAb) specific for the cytotoxic T lymphocyte-associated antigen-4 (CTLA-4, CD152) expressed on activated T cells. CTLA-4 is a negative regulator of T-cell responses. Ipilimumab augments the immune response by disrupting the interaction of CTLA-4 with its ligands, B7-1 (CD80) and B7-2 (CD86). We therefore hypothesize that ipilimumab will augment the efficacy of rituximab. This phase I with an expansion cohort was designed to assess the safety of this combined therapy. Methods: Patients ≥ 18 years old with r/r CD20+ NHL, measurable disease, KPS ≥ 70%, neutrophils ≥ 1000, platelets ≥ 50,000 and adequate kidney and liver function were eligible. In this phase I with an expansion cohort, after establishing the recommended Phase II dose (RP2D), study patients would be randomized to differing treatment schedules of rituximab and ipilimumab. For induction in phase I, rituximab was given every week for four weeks and ipilimumab was administered every 3 weeks for 4 doses. Primary endpoints included toxicity and determination of the maximum tolerated dose (MTD)/RP2D during the first 7 weeks of induction of ipilimumab (2 doses) when given in combination with rituximab (4 doses, 375mg/m2 per dose). Secondary end points included immune response, ORR and progression free survival (PFS). In the phase I component, a standard 3+3 design was used to determine the MTD/RP2D. After the RP2D was established, 20 more patients would be randomized to one of two schedules: Arm A - ipilimumab on Day 1 together with the first dose of rituximab, or Arm B - ipilimumab on Day 15 together with the third rituximab dose. After induction in both the Phase I and the expansion cohorts, ipilimumab and rituximab were given every 12 weeks (+/- 1 week) on the same day for 1 year. Results: As of August 1 2015, 21 patients have been enrolled (1SLL, 9FL, 6DLCL, 1MCL, 4 indeterminate): 15 are evaluable for toxicity assessment at 7 weeks. The median age was 62 (32-71) and the median number of prior regimens was 2 (1-6). The RP2D of ipilimumab when given with rituximab (375 mg/m2) was 3 mg/kg, which was the highest dose tested. The most common grade 1-2 AEs (>20%) were:anemia (33%); diarrhea (20%); constipation (27%); nausea/vomiting (33%); fatigue (40%); anorexia (33%); low albumin (53%); hypocalcemia (33%); hyponatremia (27%). The most common grade 3-4 AEs (≥ 5%) were: anemia (20%); diarrhea (13%); leukopenia (13%); and hyponatremia (27%). There were no study related deaths. Twelve patients remain on treatment, 6 came off due to progression, 2 due to toxicity, and there was one death (unrelated). Conclusion: The RP2D of ipilimumab when given in combination with rituximab is 3 mg/kg. This dose is safe and moderately well tolerated. Enrollment continues in the expansion cohort to assess efficacy and the ability of ipilimumab to augment the rituximab-mediated immune response. Disclosures Off Label Use: Ipilimumab to augment rituxan in treating lymphoma. Chen:Merck: Consultancy, Research Funding; genentech: Consultancy, Speakers Bureau; Millennium: Consultancy, Research Funding, Speakers Bureau; Seattle Genetics: Consultancy, Research Funding, Speakers Bureau.

Abstract Background The international prognostic score (IPS) (Hasenclever et al., NEJM 1988) uses 7 factors (age> 45, male sex, hemoglobin<10.5, stage IV, leukocytosis > 15,000, and lymphopenia< 600) to predict a 5 year freedom from progression (FFP) of 42%-84% and overall survival (OS) of 56%-89% for patients with advanced HL. Constructed from a retrospective analysis of patients treated before 1992, the IPS continues to be the most commonly used risk stratification index for advanced HL. Recent studies suggest that the predictive range of the IPS has narrowed due to improved outcomes of patients treated with current therapy (Moccia et al. JCO 2012). In this report we prospectively evaluated the ability of the individual components of the IPS to predict outcome in patients enrolled on the US Intergroup trial E2496. Methods All seven IPS (IPS-7) variables were recorded for all patients at study entry. FFP was defined as the time from study entry to disease progression or relapse; deaths that occurred during remission that were not preceded by disease progression/relapse were censored. OS was defined as the time from study entry to death from any cause. Kaplan-Meier methodology was used to construct survival curves. Univariate and multivariate analysis (MVA) was performed using Cox proportional-hazards models. We subsequently constructed an alternative prognostic score utilizing the factors which were significant on MVA (PS-3). Results From 1996-2006, 854 patients with advanced HL, were randomized to treatment with either ABVD or Stanford V, with no significant differences in outcome (Gordon et al, JCO 2013). While the IPS-7 remained prognostic it did not stratify the lowest risk patients (0-1 risk factor) or patients with 3-5 risk factors, as its predictive range was narrowed due to improved clinical outcomes (Fig 1a and 2a). Table 1 shows the univariate and multivariate analysis for IPS-7 and outcomes. In contrast to the original IPS-7, on MVA, only two factors, hemoglobin and stage were significant for FFP, and three factors for OS: hemoglobin, stage, and age. We then evaluated a new 3 factor score (PS-3) utilizing variables significant on the MVA. The PS-3 was significant for both FFP (p=0.0001) and OS (p<.0001) and clearly separated patients into 4 distinct risk groups with either: 0, 1, 2 or 3 risk factors. The five year FFP was 83%, 74%, 68%, and 63% and OS was 95%, 85%, 75%, and 52% for patients with 0, 1, 2, and 3 risk factors, respectively (Table 2, Fig 1b and 2b). Conclusion In E2496, for patients with advanced HL, the PS-3 may stratify patients more clearly compared to IPS-7 and provide a simpler model to assess risk. Studies comparing PS-3 with IPS-7, and are ongoing. The incorporation of novel biomarkers and biologic factors into PS-3, as well as validation with other data sets is warranted. Disclosures: No relevant conflicts of interest to declare.

AbstractMantle cell lymphoma (MCL) is a type of non‐Hodgkin lymphoma (NHL) with aggressive disease characteristics resulting in multiple relapses after initial treatment. Lenalidomide is an immunomodulatory agent approved in the US for patients with relapsed/refractory MCL following bortezomib based on results from 3 multicenter phase II studies (2 including relapsed/refractory aggressive NHL and 1 focusing on MCL post‐bortezomib). The purpose of this report is to provide longer follow‐up on the MCL‐001 study (follow‐ups were 6.8 [NHL‐002], 7.6 [NHL‐003], and 52.2 [MCL‐001] months). The 206 relapsed MCL patients treated with single‐agent lenalidomide (25 mg/day PO, days 1 to 21 every 28‐days) had a median age of 67 years (63% ≥65 years), 91% with stage III/IV disease, and 50% with ≥4 previous treatment regimens. With a median follow‐up of X, the combined best overall response rate (ORR) was 33% (including 11% with complete remission [CR]/CR unconfirmed CRu). Lenalidomide produced rapid and durable responses with a median time to response of 2.2 months and median duration of response (DOR) of 16.6 months (95% CI: 11.1%‐29.8%). The safety profile was consistent and manageable; myelosuppression was the most common adverse event (AE). Overall, single‐agent lenalidomide showed consistent efficacy and safety in multiple phase II studies of heavily pretreated patients with relapsed/refractory MCL, including those previously treated with bortezomib.

SummaryThe International Prognostic Score (IPS‐7) is the most commonly used risk stratification tool for advanced Hodgkin lymphoma (HL), however recent studies suggest the IPS‐7 is less discriminating due to improved outcomes with contemporary therapy. We evaluated the seven variables for IPS‐7 recorded at study entry for 854 patients enrolled on Eastern Cooperative Oncology Group 2496 trial. Univariate and multivariate Cox models were used to assess their prognostic ability for freedom from progression (FFP) and overall survival (OS). The IPS‐7 remained prognostic however its prognostic range has narrowed. On multivariate analysis, two factors (age, stage) remained significant for FFP and three factors (age, stage, haemoglobin level) for OS. An alternative prognostic index, the IPS‐3, was constructed using age, stage and haemoglobin level, which provided four distinct risk groups [FFP (P = 0·0001) and OS (P < 0·0001)]. IPS‐3 outperformed the IPS‐7 on risk prediction for both FFP and OS by model fit and discrimination criteria. Using reclassification calibration, 18% of IPS‐7 low risk patients were re‐classified as intermediate risk and 13% of IPS‐7 intermediate risk patients as low risk. For patients with advanced HL, the IPS‐3 may provide a simpler and more accurate framework for risk assessment in the modern era. Validation of these findings in other large data sets is planned.