FL: results of the phase IIIb MAGNIFY study and CIBMTR database analysis comparing auto-HCT, and MSD and MUD allo-HCT – oral abstracts at the 2017 ASCO Annual Meeting

At the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting, an oral abstract session took place that was jointly chaired by John M. Pagel, MD, PhD, of the Swedish Cancer Institute, Seattle, WA, USA, and Ranjana H. Advani, MD, from the Stanford Cancer Institute, Stanford, CA, USA.

Two key oral abstracts were presented on Follicular Lymphoma (FL):

Abstract 7502

David Jacob Andorsky, MD, from the Rocky Mountain Cancer Centers/US Oncology Research, Boulder, CO, USA, presented results of specifically of double-refractory or early relapsed patients with FL included in the phase IIIb MAGNIFY study (NCT01996865).

FL patients who experience progressive disease on immunotherapy or chemo-immunotherapy less than 24 months from initial diagnosis have abysmal prognoses. Furthermore, those patients who are refractory to chemotherapy and rituximab have very few therapeutic options available to them. Numerous reports have shown that the combination of lenalidomide and rituximab (R2) is efficacious and has a tolerable safety profile in both newly diagnosed and R/R patients with FL. In vivo, the R2 combination enhances Antibody-Dependent Cellular Cytotoxicity (ADCC) and anti-tumor effects (Wu et al. 2008; Hernandez-Ilizaliturri et al. 2005; Zhang et al. 2009). The direct effects of lenalidomide (apoptosis and cell cycle arrest) and rituximab (apoptosis) in tumor and host immune cells provides the biological basis of evaluating R2 in the MAGNIFY study.

The phase IIIb, randomized, open-label, multicenter MAGNIFY study assessed induction therapy with R2 followed by maintenance with R2 or rituximab in R/R FL, MZL, and MCL patients. The data presented by Dr Andorsky focused on FL: Double-Refractory (DR) patients are refractory to both rituximab (single-agent or in combination) and an alkylating agent; Early Relapse (ER) patients progressed or relapsed within 2 years of initial diagnosis.

Patients were administered 12 28-day cycles of lenalidomide (20mg/d days 1–21) and rituximab (375mg/m2 qwk day 1, 8, 15, and 22 of cycle 1, then day 1 every other cycle); those with SD or better are randomly assigned at a ratio of 1:1 to maintenance R2 (lenalidomide 10mg/d days 1–21 plus rituximab 375mg/m2 day 1 every other cycle) or rituximab alone (375mg/m2 day 1 every other cycle). The primary endpoint was PFS.

Of the 234 indolent NHL patients, 160 had FL with a median age of 65 years (35–91); 85 patients (53%) were aged 65 years or older. Median follow-up time for all FL patients was 10.2 months; for DR patients (n=50) this was 9.0 months and for ER patients (n=52) this was 12.1 months. Eighty-three patients were non-ER/DR. Twenty-five patients were both ER and DR. The proportion of ER, DR, and non-ER/DR patients randomized to maintenance were 23% (n=12), 18% (n=9), and 29% (n=24), respectively. The median number of previous treatments received was 2 (range, 0–9) for all patients, 3 (range, 1–8) for DR patients, and 2 (range, 1–5) for ER patients. Among ER patients, 39/52 had received first-line R-chemo and 13/52 had received rituximab monotherapy. The most common previous therapies in DR and ER patients were rituximab (26% vs. 37%), BR (66% vs. 54%), and R-CHOP/R-CHOP-like (44% vs. 37%).

The most frequently reported grade 3–4 TEAEs in DR and ER patients were neutropenia (42% vs. 37%), leukopenia (8% vs. 10%), thrombocytopenia (8% vs. 4%), and lymphopenia (6% vs. 4%). Other grade 3–4 AEs in DR and ER patients were febrile neutropenia (4% vs. 4%) and thrombosis (2% vs. 0%), no events of tumor lysis syndrome, tumor flare reaction, or hepatic disorders were reported.

Median treatment duration in all FL patients was 6.0 months, for DR patients was 5.6 months, and for ER patients was 6.2 months. Median TTR for all patients and DR patients was 2.8 months, and in ER patients was 2.7 months. Best response during induction and maintenance in efficacy evaluable patients are outlined in the table below:

Best response, n (%)

All FL (n=128)

DR (n=42)

ER (n=43)


85 (66)

19 (45)

20 (47)


49 (38)

9 (21)

9 (21)


36 (28)

10 (24)

11 (26)


31 (24)

15 (36)

17 (40)


12 (9)

8 (19)

6 (14)

In all FL patients (n=160), DR patients (n=50), and ER patients (n=52), 1-year PFS was 70%, 65%, and 49%, respectively. 1-year PFS was found to be similar in ER patients treated with front-line R-chemo (n=39; 52%) and those treated with front-line non-R-chemotherapy (n=13; 44%). Median DoR in DR and ER patients was not reached.

David Andorsky concluded the presentation by stating that induction with R2 followed by maintenance demonstrated efficacy and was well tolerated in DR or ER patients. Enrollment of the MAGNIFY trial is ongoing.

Abstract 7508

The second FL specific abstract was also focused on patients who experienced early relapse with chemoimmunotherapy. James K. Godfrey, MD, from the University of Chicago, Chicago, IL, USA, presented an analysis by the Center for International Blood and Marrow Transplant Research (CIBMTR) comparing autologous (auto) versus Matched Sibling Donor (MSD) or Matched Unrelated Donor (MUD) allogeneic (allo) Hematopoietic Cell Transplantation (HCT).

Compared to the majority of FL patients, those with high-risk disease and Early Chemoimmunotherapy Failure (ECF; relapse ≤2 years of first-line chemoimmunotherapy; occurs in approximately 20% of patients) have substantially poorer survival. Casulo et al. reported that 5-year OS in these patients is approximately 50% compared to 90% in the reference group; this was similar in their validation cohort (34% vs. 94%; IPI-adjusted HR, 19.8).

Previously, compared to conventional chemotherapy, auto-HCT has been shown to improve PFS and potentially OS in patients with relapsed FL (Schouten et al. 2003). Moreover, it has been reported by Rohatiner et al. that outcomes with auto-HCT are best when it is performed earlier in the course of the disease (≤3 lines of previous treatment).

More recently, in a study of 518 rituximab-treated patients, Klyuchnikov et al. demonstrated that allo-HCT achieves similar OS as auto-HCT (66% vs. 74%) and lowers the risk of relapse (20% vs. 54%; P < 0.0001); however, patients undergoing auto-HCT compared to allo-HCT are at a lower risk of Non-Relapse Mortality (NRM; 5-year adjusted probabilities, 5% vs. 26%; P < 0.0001).

In the analysis presented by Dr Godfrey, the CIBMTR database was used to compare auto-HCT with either MSD or MUD allo-HCT as the first transplant approach in ECF FL patients. Overall, 440 patients aged 18 years or older, who underwent auto- or allo-HCT between 2002 and 2014 and receiving first-line rituximab-based chemoimmunotherapy with evidence of ECF were included (auto = 240; MSD = 105; MUD = 95). The primary outcome measure was OS; secondary endpoints were PFS, as well as relapse mortality and NRM.


Auto (n=240)

MSD (n=105)

MUD (n=95)

Median age (range)

56 (23–79)

52 (29–68)

53 (21–74)


142 (59)

70 (67)

64 (67)

KPS ≥90%

152 (63)

71 (68)

64 (67)

Median number of therapies prior to HCT (range)

2 (1–6)

3 (1–9)

3 (1–8)

Median time from diagnosis to HCT (range), months

24 (6–203)

23 (3–128)

27 (7–167)

Myeloablative allo-HCT


35 (33)

27 (28)

Remission at HCT


86 (36)

25 (24)

19 (20)


92 (38)

32 (30)

34 (36)


55 (23)

46 (44)

40 (42)


7 (3)

2 (2)

2 (2)

Median follow-up (range), months

73 (3–142)

69 (3–152)

73 (12–121)

5-year adjusted probabilities of NRM were significantly lower with auto-HCT (5%; 2–8%) compared to MSD (17%; 10–25%) or MUD (33%; 23–43%) HCT (P < 0.0001). 5-year adjusted probabilities of relapse were significantly lower with MSD (31%; 21–40%) or MUD HCT (23%; 14–32%) versus auto-HCT (58%; 52–65%; P < 0.0001). Adjusted probabilities of 5-year PFS after auto-HCT, MSD, and MUD HCT were 38% (32–45%), 52% (41–62%), and 43% (32–54%), respectively (P = 0.006). The adjusted probabilities of 5-year OS was significantly higher following auto-HCT (70%; 64–76%) or MSD HCT (73%; 64–81%) versus MUD HCT (49%; 39–60%; P = 0.004).

On multivariate analysis, Karnofsky Score <90% (P = 0.0067) and high LDH (P = 0.0163) were associated with NRM, and stage III/IV disease (P = 0.0218) was associated with relapse/progression.

Data on causes of death was also presented:

Cause of Death

Auto-HCT (%)

MSD (%)

MUD (%)

Number of deaths




Primary Disease

67 (83)

14 (42)

20 (42)


1 (1)

6 (18)

6 (13)



5 (15)

10 (21)


13 (16)

8 (25)

12 (25)

James K. Godfrey summarized his talk by firstly stating that FL is a heterogeneous disease and ECF patients have poor outcomes (5-year OS of 34–50%). This analysis found that auto-HCT and MSD allo-HCT result in excellent 5-year OS rates of over 70%. MUD allo-HCT demonstrated good disease control but resulted in high NRM and poorer OS. Lastly, MSD allo-HCT and auto-HCT resulted in substantially higher OS versus historical controls and so present as strong contenders as treatment for ECF FL patients. 

We interviewed Prof. Gilles Salles, the chair of our Executive Steering Committee, asking what the highlights in FL were at the 2017 ASCO Annual Meeting: watch the video interview here.

  1. Andorsky D.J et al. Phase IIIb randomized study of lenalidomide plus rituximab (R2) followed by maintenance in relapsed/refractory NHL: analysis of patients with double-refractory or early relapsed follicular lymphoma (FL). J Clin Oncol 35, 2017 (suppl; abstr 7502). 2017 American Society of Clinical Oncology (ASCO) Annual Meeting; 2017 June 2–6; Chicago, IL, USA.
  2. Godfrey J.K. et al. Autologous (auto) versus matched sibling donor (MSD) or matched unrelated donor (MUD) allogeneic (allo) hematopoietic cell transplantation (HCT) in follicular lymphoma (FL) patients (pts) with early chemo-immunotherapy failure (ECF): a Center for International Blood and Marrow Transplant Research (CIBMTR) analysis. J Clin Oncol 35, 2017 (suppl; abstr 7508). 2017 American Society of Clinical Oncology (ASCO) Annual Meeting; 2017 June 2–6; Chicago, IL, USA.