We describe 20 patients with myeloma and 1 with primary amyloidosis from 15 centres, all with advanced renal failure, most of whom had PBSC mobilised using plerixafor following previous failed mobilisation by conventional means (plerixafor used up-front for 4 patients). For 15 patients, the plerixafor dose was reduced to 0.16 mg/kg/day, with a subsequent dose increase in one case to 0.24 mg/kg/day. The remaining six patients received a standard plerixafor dosage at 0.24 mg/kg/day. Scheduling of plerixafor and apheresis around dialysis was generally straightforward. Following plerixafor administration, all patients underwent apheresis. A median CD34+ cell dose of 4.6 × 106 per kg was achieved after 1 (n=7), 2 (n=10), 3 (n=3) or 4 (n=1) aphereses. Only one patient failed to achieve a sufficient cell dose for transplant: she subsequently underwent delayed re-mobilisation using G-CSF with plerixafor 0.24 mg/kg/day, resulting in a CD34+ cell dose of 2.12 × 106/kg. Sixteen patients experienced no plerixafor toxicities; five had mild-to-moderate gastrointestinal symptoms that did not prevent apheresis. Fifteen patients have progressed to autologous transplant, of whom 12 remain alive without disease progression. Two patients recovered endogenous renal function post autograft, and a third underwent successful renal transplantation. Plerixafor is highly effective in mobilising PBSC in this difficult patient group.
Clinical experience with younger myeloma patients presenting in end-stage renal failure suggests a potential benefit from autologous PBSC transplant, with 20–35% of patients recovering renal function post transplant, albeit with substantially higher early transplant-related mortality than patients with normal renal function.1, 2, 3 However, PBSC collection from myeloma patients on dialysis can be challenging, as cyclophosphamide (the most commonly used chemotherapy drug for PBSC mobilisation) is renally excreted with potentially increased toxicity in patients with renal failure.4 The alternative approach of PBSC mobilisation with G-CSF alone without prior chemotherapy has a lower success rate in terms of achieving a transplantable CD34+ cell dose.5 There is therefore a need for new approaches to PBSC mobilisation in this patient population.
Plerixafor shows considerable promise as a novel agent for PBSC re-mobilisation in patients failing to mobilise PBSC by conventional means: published experience from the US and European compassionate use programmes showed an overall success rate (in terms of achieving a transplantable PBSC dose) of approximately 70% in patients with lymphoproliferative disorders who had failed initial PBSC mobilisation by conventional means,6, 7 as opposed to historical data suggesting success rates of between 20 and 50% for second mobilisation attempts without plerixafor.8, 9, 10 However, experience of plerixafor use in the context of significant renal impairment has hitherto been limited to a single study in otherwise healthy volunteers with renal impairment.11 Specifically, one of the exclusion criteria for both the US and European compassionate use of plerixafor programmes was a serum creatinine of more than 160 μmol/litre (>1.5 mg/dL). This is because plerixafor is renally excreted,12 with theoretical potential for accumulation of the drug in patients with renal impairment. However, as a small highly polar, highly water-soluble molecule, plerixafor is also predicted to be removed efficiently by dialysis. Previous pharmacokinetic modelling in otherwise healthy volunteers with renal impairment has suggested that for patients with severe renal impairment, a daily plerixafor dose of 0.16 mg per kg should result in peak plasma levels similar to those seen in patients with normal renal function receiving a daily plerixafor dose of 0.24 mg/kg.11
Plerixafor has recently received regulatory approval in Europe and North America as an agent for autologous PBSC mobilisation. Prior to regulatory approval (licensing), the drug had been available on a named patient, Compassionate Use Programme in Europe between May 2008 and July 2009, and in the United States during 2006–2007.
We present a retrospective audit of initial European and US experience with plerixafor in myeloma patients with advanced renal failure. A series of 21 patients with myeloma and significant renal impairment at 15 centres in Europe and the United States received plerixafor for PBSC mobilisation. Twenty patients were still on dialysis at the time of PBSC mobilisation, of whom 19 were on haemodialysis and one on peritoneal dialysis. Eligible patients subsequently progressed to high-dose chemotherapy and autologous PBSC transplant.
Patients and methods
Patient characteristics & definition of study population
Patients were identified via Genzyme as having had a request for approval of plerixafor (European Compassionate Use Programme), or a request for advice regarding the use of plerixafor (post licensing in the United States or Europe) in the presence of significant renal failure (estimated glomerular filtration rate of less than 30 mL/min). Patient characteristics are shown in Table 1. Actual rather than ideal bodyweight was used for dosage calculations.
Chemotherapy before PBSC mobilisation
Details are shown in Table 1. This reflects considerable diversity of practise regarding initial treatment of younger myeloma patients within Europe and the United States of America. Fifteen patients showed satisfactory response to first-line induction therapy and progressed to PBSC collection thereafter; six patients switched to second-line induction therapy because of poor disease response or thalidomide neurotoxicity.
PBSC mobilisation attempts before plerixafor mobilisation
These are summarised in Table 2. There were 21 mobilisation attempts in 17 patients before plerixafor (4 patients underwent first-line plerixafor mobilisation), of which 16 attempts were completely unsuccessful with no apheresis carried out because of inadequate peripheral CD34+ cell counts. The remaining five mobilisation attempts yielded an insufficient or suboptimal CD34+ cell dose for transplant according to local institutional criteria (median CD34+ cell dose 0.8 × 106/kg; range 0.43–2.5)—two of these patients achieved a CD34+ cell dose greater than 2.0 but less than 2.5 × 106/kg, with local institutional minimum for autologous transplant being 2.5 × 106/kg.
Re-mobilisation of HSC using plerixafor
None of the patients underwent back-up autologous BM harvest. For 12 of the 17 patients failing initial PBSC mobilisation, and for the 4 patients who underwent first-line plerixafor mobilisation, a standard administration schedule was used after a short course of G-CSF, as previously published.6 Plerixafor (dosage discussed below) was given in the evening following 4 days of G-CSF 10 μg/kg/day, with a further dose of G-CSF 10 μg per kg on the morning of day 5, at least 1h before apheresis. Apheresis was commenced on day 5 of G-CSF at approximately 11 h post plerixafor (apheresis day 1). If a transplantable PBSC dose was not achieved on apheresis day 1, plerixafor was administered that evening with further G-CSF and apheresis the following morning; if a transplantable dose was still not achieved, plerixafor was administered on the evening of apheresis day 2 with further G-CSF and apheresis the following morning. A single patient received plerixafor on the evening of apheresis day 3 with a fourth apheresis procedure the following morning.
For the remaining five re-mobilised patients, plerixafor was used ‘pre-emptively’ after chemotherapy, as previously published,13 starting at the time of WBC regeneration following IV cyclophosphamide 2 g/m2 with G-CSF 10 μg/kg/day (G-CSF started on day +1 post chemotherapy). Apheresis was carried out approximately 10–11 h after each dose of plerixafor.
Six patients received plerixafor at the full standard dosage of 0.24 mg/kg/day. On the basis of a subsequent pharmacokinetic modelling study in otherwise healthy volunteers with renal impairment,11 the remaining 15 patients received plerixafor initially at reduced dosage of 0.16 mg/kg/day; this dose was increased for one patient from apheresis day 2 onwards to 0.24 mg/kg per day because of poor response.
Scheduling of plerixafor and apheresis around dialysis sessions
This was potentially problematic, but a workable programme was drawn up as shown in Figure 1, to allow both apheresis and dialysis sessions to happen within office hours in a standard working week. Plerixafor is predicted to be removed by haemodialysis, and the schedule below allows this to be done within a maximum of 48 h for each dose. This schedule was used for 13 patients; one of the remaining dialysis-dependent patients mobilised with G-CSF plus plerixafor had additional haemodialysis immediately following apheresis over and above their usual thrice-weekly schedule.
The remaining patient mobilised with G-CSF plus plerixafor was on peritoneal dialysis rather than haemodialysis. Peritoneal dialysis was discontinued before each dose of plerixafor (given in the evening), and recommenced immediately after completion of apheresis the following day.
For the five patients where plerixafor was used pre-emptively during regeneration from cyclophosphamide 2 g/m2 given as mobilising chemotherapy, scheduling proved more problematic, with either unscheduled dialysis sessions or missed apheresis days in all cases.
One patient, though on haemodialysis at presentation, was no longer dialysis-dependent at the time of mobilisation with plerixafor, though renal function remained poor with eGFR of 20 mL/min.
PBSC mobilisation outcomes
All 21 patients ultimately mobilised a transplantable PBSC dose using plerixafor, of whom 20 were successful at the first attempt, whereas one patient did not initially mobilise a transplantable CD34+ cell dose, but underwent successful repeat mobilisation with plerixafor plus G-CSF approximately 6 weeks later. Mobilisation outcomes for all 21 patients following first plerixafor mobilisation are summarised in Table 3. Median peak CD34+ cell count achieved was 33.3/μL. No significant difference was seen between the subgroup of patients receiving plerixafor 0.24 mg/kg/day (n=6; median peak CD34+ cell count=34) and the subgroup receiving plerixafor 0.16 mg/kg/day (n=15; median peak CD34+ count=33.3; P=NS).
The two patients who had achieved a CD34+ cell dose between 2.0 and 2.5 × 106 per kg in mobilisation attempts before plerixafor, and who might therefore be regarded as ‘suboptimal’ rather than ‘failed’ mobilisers, both mobilised well with plerixafor, attaining a CD34+ cell dose of 6.1 × 106 per kg in one apheresis, and 3.1 × 106 per kg in two aphereses, respectively.
The patient on peritoneal dialysis successfully mobilised a CD34+ cell dose of 3.01 × 106 per kg in two aphereses (two plerixafor doses), without plerixafor toxicity.
The patient with poor renal function but not on dialysis at the time of plerixafor mobilisation received three doses of plerixafor 160 μg/kg, with three aphereses resulting in a total CD34+ cell dose of 2.3 × 106/kg, without plerixafor toxicity.
The one patient initially failing plerixafor mobilisation was re-mobilised with G-CSF 10 μg/kg/day and plerixafor 240 μg/kg/day; the total CD34+ cell dose achieved at re-mobilisation was 2.12 × 106/kg.
Sixteen patients did not suffer any apparent plerixafor toxicity. Five patients suffered mild-to-moderate gastrointestinal toxicity, National Cancer Institute Grade 2 or less (see Table 4). Following antiemetic therapy for nausea or vomiting, and antidiarrhoeal therapy where necessary, all patients were able to undergo apheresis each morning after plerixafor administration.
Fifteen (71%) patients proceeded to autologous transplant. Conditioning was with high-dose melphalan. The median transplant CD34+ cell dose was 5.3 × 106/kg (range 2.3–7.56). In all 15 cases, the transplanted cells had been collected entirely using plerixafor, that is, cells were not pooled with cells from previous partially successful mobilisation attempts before plerixafor. For four patients, only half of the cells collected using plerixafor were actually transplanted, with the remainder being held in reserve for potential delayed second transplant.
One patient died of sepsis at day +33, although engraftment was achieved; two patients engrafted but experienced disease progression, eventually leading to death at 10 months and 15 months post transplant. The remaining 12 transplanted patients are alive without disease progression at a median of 7.5 months from transplant (median OS and disease-free survival not yet reached). One of these patients underwent an immediate second autologous transplant (i.e. a planned up-front tandem autologous transplant) using PBSC previously collected with plerixafor. Of the 14 patients who were still dialysis-dependent at the time of PBSC transplant, two recovered endogenous renal function post transplant; a third achieved CR from myeloma and underwent a successful renal transplant.
All 15 transplanted patients achieved neutrophil and platelet engraftment. Median time to neutrophils >0.5 × 109/L was 11.5 days (range 9–14 days); median time to platelets >20 × 109/L (first of three consecutive days with platelets >20 × 109/L without transfusional support) was 15 days (range 10–33 days). One patient experienced delayed platelet engraftment (33 days). There were no secondary graft failures.
Six patients have not proceeded to transplant. Two patients died of sepsis (pneumonia) at 75 days and 90 days post plerixafor, without undergoing transplant. One patient was not transplanted because of disease progression (n=1; subsequently responded to bortezomib), but remains alive. The remaining three patients are alive with stable disease, but have elected not to proceed to transplant until the time of disease progression.
Our results in this initial series of 21 myeloma patients presenting with dialysis-dependent renal failure confirm plerixafor as a promising agent for this patient group, with all patients ultimately mobilising a transplantable cell dose after one or (in one case) two mobilisation episodes using plerixafor, with acceptable toxicity, and satisfactory engraftment for the 15 patients progressing to autologous PBSC transplant.
The initial mobilisation success rate with plerixafor for the 17 patients in this series who had previously failed PBSC mobilisation was, at 94%, higher than that described in the published experience with non-dialysis myeloma patients in the US compassionate use programme, which was just 71.4%.6 We speculate that the apparent higher success rate experienced with patients with dialysis-dependent renal failure may reflect delayed excretion of the drug with a more prolonged effect on circulating CD34+ cell counts. However, difference in apheresis practise or other confounding variables between our patients and the published compassionate use experience6, 7 cannot be ruled out. It should perhaps be noted that only one patient in this series had received lenalidomide before mobilisation with plerixafor: lenalidomide therapy is a recognised risk factor for poor PBSC mobilisation,14 and while it has hitherto been little used in patients with significant renal impairment because of concerns regarding increased toxicity, recent positive experience with dose-reduced lenalidomide regimes in this patient group15 may potentially lead to increasing use in the future.
Reassuringly, despite the likelihood of delayed plerixafor excretion in the presence of significant renal failure, toxicity was acceptable in this patient group, being limited to moderate gastrointestinal toxicity in five patients (see Table 4). No non-gastrointestinal toxicities (bone pain, injection-site reactions and so on) were reported.
The optimum dosage of plerixafor in patients with advanced renal failure remains to be determined. Patients in our series received either standard dosage of 240 μg/kg/day (n=6) or (initially) reduced dosage of 160 μg/kg/day (n=15). No significant difference was observed either in peak peripheral CD34+ cell count attained (see Table 3) or in occurrence of gastrointestinal toxicity (see Table 4) between the two dosage subgroups: however, it is not possible to comment further because of low patient numbers.
We have also demonstrated for 13 of the patients in our series (all re-mobilised with G-CSF plus plerixafor without mobilising chemotherapy) that a simple schedule for the timing of plerixafor, G-CSF and apheresis around thrice-weekly haemodialysis sessions allows myeloma patients on regular haemodialysis to remain on their usual dialysis schedule while still allowing apheresis to be completed within a standard working week. This has potential advantages over chemotherapy-based mobilisation approaches (+/− pre-emptive plerixafor) for this patient group, where the timing of apheresis was found to be less predictable for the five such patients in this series, and less straightforward to accommodate around scheduled haemodialysis sessions.
We have shown mobilisation of PBSC for transplant using plerixafor to be a safe and feasible option for myeloma patients with advanced renal failure, with an initial 95% success rate (over the whole series of 21 patients) in mobilising a transplantable CD34+ dose after a median of two doses of plerixafor (approximate current cost EUR12 000 or US $13 00016) in a series of 21 patients. Although formal cost-effectiveness analysis has not been carried out because of relatively low patient numbers, there are potential advantages in terms of predictability of apheresis scheduling and avoidance of in-patient admission for mobilising chemotherapy that are likely to offset much of the additional cost involved.17 Importantly, given that 17 of these patients had previously failed PBSC mobilisation by conventional means, re-mobilisation with plerixafor appears to have a considerably higher success rate than the 20–50% success rates in achieving a CD34+ cell dose of >2 × 106 per kg reported for alternative re-mobilisation strategies,8, 9, 10 and despite the fact that the drug is renally excreted, we only observed mild-to-moderate gastrointestinal toxicities in a minority of patients. Given the considerable efficacy and low toxicity of plerixafor in myeloma patients with advanced renal failure who have failed previous PBSC mobilisation, there may now be grounds for evaluating its role as first-line PBSC mobilisation therapy in this patient group.
We thank the patients, nurses, data managers and stem cell laboratory staff at all participating centres. We would also like specifically to thank the following individuals for their assistance: Kristen Kemp, Program Coordinator & Quality Management Supervisor, Aurora St Luke's Medical Center, Milwaukee; Marie Ventimiglia, Bone Marrow Transplant Clinical Research Coordinator, Karmanos Cancer Institute, Detroit; Juli Murphy, Manager, BMT Clinical Research, Rocky Mountain Blood and Marrow Transplant Program, Denver; Dr Tharani Balasubramaniam, Leeds Teaching Hospitals, U.K.; Joy Sinclair, Apheresis Nurse Manager, Glasgow, U.K.; Dr Nicky Whitaker, Genzyme Europe; Purvi Mody, Genzyme Corporation USA; Pritesh Gandhi, Genzyme Corporation USA.