PACIFICO

• Research type

  Research Study

• Full title

  Purine-Alkylator Combination In Follicular lymphoma Immuno-Chemotherapy for Older patients: a phase III comparison of first-line R-CVP versus R-FC

• IRAS ID

  13686

• Contact name

• Eudract number

  2008-004759-31

• ISRCTN Number

  ISRCTN99217456

• Research summary

  This study is a two arm, open-label, multi-centre, parallel group, phase III, randomised controlled trial comparing rituximabfluarabine, cyclophosphamide(R-FC) and rituximab, cyclophosphamide, vincristine, prednisolone(R-CVP)in older patients with symptomatic advanced-stage follicular lymphoma. 680 patients will be randomised in total, 340 into each arm. The primary aim of the trial will be to compare two different induction regimens in order to determine which achieves the best balance between efficacy and toxicity. Patients will receive 24 weeks of treatment and responders will then go on to receive 2 years of rituximab maintenance. All patients will be followed up for a minimum of 3 years from trial entry.The schedule of treatment is as follows:Arm 1: R-CVP repeated every 21 days for 8 cyclesArm 2: R-FC repeated every 21 days for 4 cycles followed by rituximab alone for 4 further cycles.Rituximab maintenance: All responding patients will receive rituximab maintenance (2 monthly for 2 years).Quality of life and Health Economic questionnaires (EORTC QLQ C-30,EQ-5D,EQ-VAS) will be completed at the baseline, 2 weeks after the 4th and 8th cycles during induction therapy and every 4 months thereafter.The trial will support a wide range of laboratory studies aiming to elucidate novel biomarkers that predict therapeutic response or toxicity. To facilitate these studies, the following biosamples will be collected at baseline and stored centrally: a) Formalin fixed paraffin embedded (FFPE) lymph node biopsies b) Anti-coagulated blood c) Clotted blood for serum studiesIn addition, sequential anti-coagulated and clotted blood samples may be collected from selected patients for dynamic biomarker studies. Collecting these additional samples would not involve additional study visits or venesections. If additional samples are required procedures will be included as a protocol amendment

  Research Summary:
  This phase III randomised controlled trial (RCT) compared two different chemoimmunotherapy induction regimens in older people with previously untreated, symptomatic, advanced-stage follicular lymphoma. The standard arm consisted of 8 cycles of rituximab, cyclophosphamide, vincristine and prednisolone (R-CVP), given every 3 weeks. The experimental arm comprised 4 cycles of rituximab, fludarabine and cyclophosphamide (R-FC) followed by 4 cycles of rituximab alone, given every 4 weeks. Patients who achieved a complete or partial response then received rituximab maintenance every 8 weeks for up to 12 doses. Two key protocol changes were made during the course of the study: (1) rituximab was switched from the intravenous to subcutaneous route of administration (except for the first dose); and (2) the fludarabine and cyclophosphamide components of the R-FC regimen were spread over 4 rather than 3 days with an overall dose reduction of 17.5% and 36%, respectively. Prophylaxis with co-trimoxazole (or equivalent) was mandated in all patients receiving R-FC, together with acyclovir in those with a prior episode of herpes zoster. Patients were followed up for a minimum of 6 months after completion of all treatment, with assessments for response at the end of induction and for progression, toxicity and quality-of-life (QoL) throughout the study. Reporting of serious adverse events (SAEs) was mandated until 28 days after the last day of treatment, except for grade 3-4 infections and neoplasms where reporting requirements were extended to 6 months post treatment and the end of follow-up, respectively.
  369 patients were enrolled at 80 UK sites between October 2009 and April 2016, with 185 randomised to R-FC and 184 to R-CVP. Within the R-FC group, 53 patients were allocated to full-dose R-FC (fdR-FC) and 132 to dose-attenuated R-FC (daR-FC). The R-CVP, fdR-FC and daR-FC cohorts were well balanced for age (median 75, 75 and 76.5), sex (female in 54%, 53% and 55%), FLIPI (high in 64%, 68% and 64%) and co-morbidity (CIRS >6 in 19%, 17% and 20%). However, more patients in the fdR-FC group had a WHO performance score of 0 (fully fit) compared to the daR-FC group (64% vs 36%), illustrating a shift in the profile of enrolled patients towards lower fitness levels following the switch from fdR-FC to daR-FC. Following randomisation, one patient was found to be ineligible, and another withdrew consent, leaving 367 patients in the intention-to-treat (ITT) population.
  95%, 92% and 89% of patients in R-CVP, fdR-FC and daR-FC cohorts completed the first 4 cycles of induction, while 83%, 74% and 76% completed all 8 cycles. The mean (SD) number of induction cycles administered in the respective cohorts was 7.4 (1.6), 6.9 (1.9) and 7.1 (1.9), while the total number of cycles administered (including maintenance) was 16.4 (5.7), 14.5 (6.5) and 14.9 (6.1).
  The co-primary outcomes were progression-free survival (PFS) and grade 3-4 infections. The probability of being alive and progression free among all 369 patients randomised to R-FC vs R-CVP was 71% vs 72% at 4 years and 61% vs 42% at 8 years, with KM curves diverging after ~4 years. However, the difference did not reach statistical significance irrespective of whether a non-stratified [HR 0.80 (95% CI: 0.58, 1.11); P 0.188] or stratified [0.65 (0.40, 1.05); P=0.077] Cox PH model was used. Since there was evidence of non-proportionality in the Cox model, further analysis was performed using Restricted Mean Survival Time (RMST). However, there was no difference in the RMST ratio [1.05 (95% CI: 0.95, 1.17)].
  Subgroup analysis showed broadly similar findings in patients with low, intermediate or high FLIPI scores, and in those with more (CIRS >6) or less (CIRS ≤6) comorbidity. Notably, separation of PFS curves for R-FC vs R-CVP was only evident among the 261 patients who were randomised after the switch from fdR-FC to daR-FC [HR 0.7 (95% CI: 0.46, 1.05); P=0.08], with no discernible difference among the 103 patients who were randomised before this switch [1.03 (95% CI: 0.59, 1.82); P=0.911].
  There was no significant difference in the proportion of patients experiencing at least one grade 3-4 infection [risk ratio: 1.11 (95% CI: 0.80, 1.55); P=0.617] or in the occurrence of such infections [odds ratio: 1.16 (95% CI: 0.73, 1.83)]. Subgroup analysis showed broadly similar findings in patients who were randomised before or after the switch from fdR-FC to daR-FC, in those with low, intermediate or high FLIPI scores, and in those with more (CIRS >6) or less (CIRS ≤6) comorbidity.
  Secondary outcomes included overall survival (OS), time to next treatment (TNT), high-grade transformation (HGT), anatomical response, response duration and analysis of frailty and comorbidity. There was no difference in OS between R-FC and R-CVP [HR: 1.04 (95% CI: 0.71, 1.52); P=0.842], the probability of remaining alive being 85% vs 78% at 4 years and 65% vs 60% at 8 years. In contrast, TNT was significantly longer for R-FC compared to R-CVP [HR 0.5 (95% CI: 0.33, 0.71); P=0.001], the probability of remaining free of second-line lymphoma treatment being 84% vs 72% at 4 years and 80% vs 58% at 8 years. Rates of HGT were similar for R-FC (5%) and R-CVP (4%).
  There was a significant difference in PFS between R-FC and R-CVP by patients classified as ≥3 in the Vulnerable Elders Survey [HR: 0.28 (95% CI: 0.09, 0.86); P=0.018] and ≥4 in the Groningen Frailty Index [HR: 0.47 (95% CI: 0.23, 0.94); P=0.028].
  Among the 298 patients who were evaluable for response, overall response (OR) rates for R-FC and R-CVP were 97% vs 99%, respectively, with complete response (CR) rates of 42% vs 39%. Response duration was significantly longer for R-FC compared to R-CVP [HR 0.48 (95% CI: 0.29, 0.79); P=0.004], the probability of remaining progression free being 89% vs 79% at 4 years and 86% vs 65% at 8 years.
  Among the 366 patients who received at least one treatment dose, more SAEs of all grades were reported following R-FC compared to R-CVP (243 vs 178). SAE rates for fdR-FC, daR-FC CVP and R-CVP were 1.5, 1.3 and 1.0, respectively, with fatal SAEs occurring in 5/52 (9.6%), 4/130 (3.1%) and 4/178 (2.2%) patients, respectively. The most common SAEs in patients receiving R-FC or R-CVP were infections (102 vs 71) and neoplasms (35 vs 16) which together accounted for 53% of all SAEs. The most common infections were lower respiratory (43 vs 22) and neutropenic sepsis (18 vs 16), while the most common neoplasms were non-melanoma skin cancers (16 vs 1). Regarding SAEs of special interest, there were three reports of Pneumocystis jirovecii pneumonia (one each with fdR-FC, daR-FC and R-CVP), two of cytomegalovirus reactivation (both with fdR-FC), and five of myelodysplasia (three with fdR-FC, one each with daR-FC and R-CVP). Regarding supportive care, red-cell transfusions were administered to 3% patients in the R-CVP cohort, 13% in the fdR-FC cohort and 8% in the daR-FC cohort. The corresponding rates for immunoglobulin infusions were 1%, 6% and 5%, while those for granulocyte colony stimulating factor (G-CSF) administration were 21%, 34% and 34%.
  QoL scores obtained with EORTC QLQ-C30, EQ-5D-3L and EQ-VAS questionnaires were similar for R-CVP and R-FC. This suggests that any reduction in QoL due to the higher toxicity of R-FC compared to R-CVP was balanced by an improvement in QoL due to the longer remissions obtained.
  Conclusion: R-FC was superior to R-CVP in terms of TNT, response duration and long-term QoL. R-FC was also associated with greater supportive care requirements and more SAEs, especially lower respiratory infections and non-melanoma skin cancers. Despite having a lower level of fitness, patients in the daR-FC cohort experienced significantly fewer SAEs compared to the fdR-FC cohort, with a fatal SAE rate only 1% higher than for R-CVP. Failure of the longer TNT associated with R-FC to translate into a statistically significant PFS advantage may be explained by the relatively high background death rate in this elderly population with a commensurate narrowing of the difference in proportional hazard. This, together with the effectiveness of subsequent therapies, may also explain the lack of a demonstrable OS advantage. In summary, daR-FC is a highly effective treatment option in follicular lymphoma with an acceptable safety profile, even in an elderly population. The flattening of KM plots for TNT and response duration after ~5 years is unexpected and indicates that responses induced by R-FC may be extremely durable, potentially amounting to a function cure, in a significant proportion of patients.
  

• REC name

  North West - Liverpool Central Research Ethics Committee

• REC reference

  09/H1005/29

• Date of REC Opinion

  19 Jun 2009

• REC opinion

  Further Information Favourable Opinion