High activity 90Y-ibritumomab tiuxetan (Zevalin®) with peripheral blood progenitor cells support in patients with refractory/resistant B-cell non-Hodgkin lymphomas

Authors


Pier Francesco Ferrucci, Melanoma and Sarcoma Division, Clinical Translational Research Program, European Institute of Oncology, Via Ripamonti, 435, 20100 Milan, Italy. E-mail: pier.ferrucci@ieo.it

Summary

Radioimmunotherapy (RIT) is an alternative approach in the treatment of resistant/refractory B-cell non-Hodgkin lymphoma (NHL). We performed a feasibility and toxicity pilot study of escalating activity of 90Y-ibritumomab tiuxetan followed by autologous stem cell transplantation (ASCT). Three activity levels were fixed – 30 MBq/kg (0·8 mCi/kg), 45 MBq/kg (1·2 mCi/kg) and 56 MBq/kg (1·5 mCi/kg) – and 13 patients enrolled. One week before treatment all patients underwent dosimetry. ASCT was performed 13 d after Zevalin® administration. Treatment was well tolerated and all patients engrafted promptly. No differences in terms of haematological toxicities were observed among the three levels, apart from a delayed platelet recovery in heavily pretreated patients receiving 56 MBq/kg. Non-haematologic toxicity was mainly related to infections and liver toxicity. One patient died 4 months after treatment because of hepatitis C virus reactivation. One patient developed a myelodysplastic syndrome 2 years after treatment. In conclusion, high-activity Zevalin® with ASCT is feasible and could be safely delivered in elderly and heavily pretreated NHL patients, including those who previously received high-dose chemotherapy and ASCT. Maximum tolerated dose was not clearly defined according to dosimetry and clinical toxicities, and further studies are needed to confirm the toxicity profile and evaluate efficacy.

Sequential high-dose chemotherapy (HD-CT) with autologous stem cell transplantation (ASCT) is a widely accepted procedure in the treatment of patients affected by relapsing B-cell non-Hodgkin lymphoma (NHL) (Vose et al, 1993; Philip et al, 1995; Nademanee et al, 2000); however, it is precluded to elderly patients because of treatment-related morbidity and mortality. For those patients who are not suitable for HD-CT, therapeutic options are limited and less effective.

Rituximab – a chimeric anti-CD20 monoclonal antibody (Moab) – has been found to be effective in untreated and relapsed B-cell NHL, alone or in association with CT (Maloney et al, 1994, 1997a,b; McLaughlin et al, 1998; Foran et al, 2000; Vose et al, 2001; Coiffier et al, 2002). Conjugation with radio-nuclides represents a valid option to strengthen the efficacy of Moab supplying doses of radioactivity to specific tumoral targets (Wiseman et al, 2003).

Yttrium-90 (90Y)-ibritumomab tiuxetan (Zevalin®, Idec Pharmaceuticals Corporation, San Diego, CA, USA) is a new anti-CD20 antibody that includes Ibritumomab, a murine parent of the humanized anti-CD20 Moab Rituximab, conjugated by Tiuxetan to 90Y. It has been shown to be active and safe in the treatment of follicular lymphoma (FL) and diffuse large B-cell lymphoma (DLBCL), even in elderly patients resistant or refractory to Rituximab (Witzig et al, 1999, 2002a,b; Morschhauser et al, 2007). Response duration, however, is still considered too short (Morschhauser et al, 2007).

Despite the absence of randomized studies, phase I and II experiences showed limited toxicity and better response rates (RR) in terms of complete remissions (CR), disease-free (DF) intervals and overall survival (OS) when 131Iodine (131I) was administered at elevated activity supported by ASCT (Press et al, 1995; Liu et al, 1998).

High acitivity (HA) 90Y-ibritumomab tiuxetan has been delivered only in combination with HD-CT, demonstrating that it can be administered safely, with no significant increase in transplant-related toxicities when rescued by ASCT (Press et al, 1993; Juweid et al, 2000; Gopal et al, 2001, 2005; Winter et al, 2001, 2004; Devizzi et al, 2005; Krishnan et al, 2005; Nademanee et al, 2005; Shimoni et al, 2005).

There is no extended information regarding the myeloablative activity of 90Y-ibritumomab-tiuxetan administered alone with haematological support in the treatment of NHL, either on the best timing for ASC-reinfusion or on the clinical efficacy and safety of this procedure in that context.

Based on these observations, 13 patients affected by resistant/relapsed NHL and unsuitable to receive HD-CT approach, because of age, developed co-morbidities or previous transplantation, were enrolled in a monocentric pilot study to determine the feasibility and safety of escalating 90Y-ibritumomab tiuxetan when administered with ASC support.

Materials and methods

Eligibility

All patients had histologically confirmed, refractory or transformed CD-20 positive B-cell NHL. They were considered unsuitable to receive HD-CT when >65 years old or in case of relapse or progression after a previous autologous transplantation. Patients were >18 years old, had adequate cardiac (cardiac ejection fraction >50% by echocardiogram), pulmonary [forced expiratory volume (FEV1) >65% of predicted or a carbon monoxide diffusing capacity (DLCO) ≥50% of predicted], renal and liver functions (serum creatinine level <176·8 μmol/l, total bilirubin level <42·75 μmol/l, aspartate transaminase/alanine transaminase level <4 times normal), <25% bone marrow (BM) involvement, a performance status of 0, 1, or 2 according to the World Health Organization Scale and a life expectancy of at least 3 months.

At time of treatment, the platelet count (PLT) should have been ≥100 × 109/l, while no limit was fixed for white blood cell (WBC) count and haemoglobin (Hb) because of subsequent ASC reinfusion.

Patients were excluded if they had received cytotoxic CT, radiotherapy or cytokine therapy within the previous 4 weeks, as well as prior radioimmunotherapy (RIT) or if they had a history of human anti-murine antibodies (HAMA) or human anti-chimeric antibodies (HACA). Previous HD-CT with ASCT, bulky disease or prior exposure to Rituximab were not considered as exclusion criteria. The study was approved by the Ethics Institutional Review Board and written informed consent was obtained from all patients.

Study design

This was a pilot study to establish the feasibility and toxicity of HA 90Y-ibritumomab tiuxetan followed by ASCT. RR, duration of responses and OS were secondary end-points. Three escalating activity levels were planned to verify and compare the safety profile of the procedure: 30 MBq/kg (0·8 mCi/kg), 45 MBq/kg (1·2 mCi/kg) and 56 MBq/kg (1·5 mCi/kg). Four patients were planned at each activity-level. Dosimetry was performed in all patients. No patient was enrolled in a higher level before verifying engraftment of the previously treated ones.

Radiological – computed tomography scan (CTS) of the neck, chest, abdomen, and pelvis and BM evaluation were performed within 4 weeks prior to treatment. All patients underwent cardiologic and pulmonary evaluation at baseline and then 4 weeks, 6 months and 1 year after 90Y-ibritumomab tiuxetan. Use of prophylactic growth factors was not recommended, while all patients received antibiotic prophylaxis during neutropenia.

During the treatment, blood counts (complete blood count/differential/platelets) were collected on day −7, 0, 7, 13; then daily until recovery (absolute neutrophil count, ANC > 1·0 × 109/l and PLT > 20·0 × 109/l), weekly until the end of treatment period and every 2 months for the first year. BM aspiration was planned annually to monitor for incoming myelodysplastic syndrome.

Drug administration and dosimetry

The ASC were collected via aphaeresis after mobilization with growth factors (granulocyte colony-stimulating factor: GCSF) alone or together with chemotherapy (cyclophosphamide 2 g/m2), having a target of at least 2·0 × 109 CD34 + /kg.

Patients received 111In-ibritumomab tiuxetan on day −20 and a therapeutic dose of 90Y-ibritumomab tiuxetan on day −13. Both doses were preceded by an infusion of the unlabelled antibody Rituximab (250 mg/m2) to deplete B cells from the peripheral circulation, BM and lymph-nodes, optimizing 90Y-ibritumomab-tiuxetan bio-distribution (Fig 1).

Figure 1.

 Study design. Patients received 111In-Zevalin® on day −20 followed by a therapeutic activity of 90Y-ibritumomab tiuxetan on day −13. Both administrations were preceded by an infusion of the unlabelled antibody Rituximab (250 mg/m2), to optimized bio-distribution. Following administration of 111In-Zevalin®, serial anterior/posterior whole body scans were acquired at 0, 1, 16, 24, 96 and 144 h after infusion to evaluate the distribution of activity in critical organs. Urine and blood samples were also collected at the same timing. Reinfusion was performed on day 0. CT, computed tomography scan; BM, bone marrow; SQM, square metre; PLT, platelet count; ANC, absolute neutrophil count.

Patient-specific dosimetry was performed in all cases. Following administration of 111In-ibritumomab tiuxetan, serial anterior/posterior whole body scans were acquired at 0, 1, 16, 24, 96 and 144 h after infusion to evaluate the distribution of activity in critical organs. Blood and urine samples were also collected according to the same timing. Data were analysed according to the medical internal radiation dose method, using the conjugate view technique, to obtain predicted absorbed doses to non-disease-involved organs (Cremonesi et al, 2000). The absorbed doses were calculated using Organ Level Internal Dose Assessment, including individual organ masses.

Treatments were administered on an inpatients basis, in accordance with local regulations. On day 0, ASC were reinfused and patients remain hospitalized until recovery from all haematological and non-haematological toxicities. Engraftment was considered delayed if, on day 28 from reinfusion, the ANC was <1·0 × 109/l or PLT were not ≥20·0 × 109/l.

Toxicity assessment

Adverse events were assessed according to the Common Toxicities Criteria of the National Cancer Institute (NCI, version 3.0).

HAMA assessment

Blood samples for HAMA analysis were collected at baseline, at 2 and 6 months from therapy and yearly thereafter. After collection, blood samples were centrifuged and sera were harvested and stored at −80°C until analysis. Detection was performed by a remote centralized laboratory using a validated protocol based on an enzyme-linked immunosorbent assay (HAMA ELISA Kit, Cat. N. 1 985 493, Roche Diagnostics, Mannheim, Germany).

Response assessment

The first response-evaluation was performed 2 months after 90Y-ibritumomab-tiuxetan administration, and repeated every 2 months until 6 months from treatment and thereafter every 4 months until disease progression or death. Disease status was determined by physical examination, CTS of the neck, chest, abdomen and pelvis and BM biopsy, if it was positive for lymphoma at baseline.

Complete response was defined as complete resolution of all disease-related radiological abnormalities and the disappearance of all signs and symptoms related to disease. Patients with BM involvement at baseline were required to repeat BM biopsy after therapy to confirm a CR. Partial response (PR) was defined as a reduction by at least 50% in the sum of the products of the largest perpendicular diameters of all measurable lesions. Progressive disease (PD) was defined as an increase of at least 25% in the sum of the products of the largest perpendicular diameters, or a new lesion larger than 2 cm as revealed by radiography or 1 cm according to physical examination, or involvement of the BM in a patient who had reached a previous CR or a clinical CR. Patients were considered to have stable disease if they did not meet criteria for CR, PR or PD.

Statistics

Demographics and other patient characteristics at study entry are summarized in Table I. Summary statistics (number, mean, standard deviation, median and 95% confidence intervals) for time to nadir by dose level for PLT and ANC are shown in Table II. Differences among dose level for time to nadir were analysed using a log-rank test. Scatter plots for WBC, platelets, neutrophils and Hb by time are showed in Fig 2A–C.

Table I.   Patient characteristics and disease status at study entry.
FactorAll patient (n)Level 1 (n)Level 2 (n)Level 3 (n)
  1. Activity levels 1, 2 and 3 correspond to 30 MBq/kg, 45 MBq/kg and 56 MBq/kg respectively. Age, previously reported toxicity and chemorefractoriness (two prior regimens or anthracycline if low-grade disease) suggest that HD CT induction is not indicated for autologous transplantation.

  2. LDH, lactate dehydrogenase (normal level: 240–480 UI/l; low level: <240 UI/l; high level: >480 UI/l); IPI, International Prognostic Index; MCL, mantle-cell lymphoma; DLBCL, diffuse large B-cell lymphoma.

Age, years
  Mean60·6496268·8
  Median684866·571
  Range28–7328–7242–7363–71
Age group
  <60 years3210
  >60 years10235
Sex
  Male11434
  Female2011
Stage at diagnosis
  I/II1001
  III/IV12444
Pathology report histology
  MCL3120
  DLBCL8314
  Follicular1010
  Transformed1001
Bone marrow involvement
  0%13445
  0·1–25%0000
Bulky disease
  <7 cm11335
  >7 cm2110
Baseline LDH
  Normal or low10235
  High3210
Performance status
  0, 112345
  21100
Months from diagnosis to treatment
  Median 28184420
  Range9–499–44 28–4911–40
No. of prior regimens
  Median342·52
  Range2–62–52–42–6
Prior radiotherapy
  Yes3111
  No10334
IPI risk group at study entry
  Low/intermediate 7223
  Intermediate/high6222
Status of disease at treatment
  Complete remission2101
  Partial remission6132
  Stable disease2101
  Progressive disease3111
Table II.   Absorbed doses to target organs in relation to activity administered: median values and ranges of variability (Gy).
Target organsAbsorbed doses (Gy) median values and ranges of variability
Level 1Level 2Level 3
  1. Activity levels 1, 2 and 3 correspond to 30 MBq/kg, 45 MBq/kg and 56 MBq/kg respectively.

Heart wall5·8 (5·1–7·5)6·6 (4·5–10·4)9·7 (6·1–19·9)
Lungs3·2 (2·5–3·8)5·2 (1·0–8·4)7·5 (5·5–8·4)
Liver7·2 (5·1–12·3)11·7 (7·9–31)13·0 (9·5–25·3)
Spleen4·0 (2·1–6·9)10·7 (3·6–14·7)7·7 (5·3–14·9)
Kidneys3·0 (1·1–7·7)5·9 (2·7–8·4)6·5 (3·3–11·9)
Red marrow1·9 (0·9–2·0)2·3 (1·8–3·3)3·7 (2·8–4·8)
Testes3·2 (2·4–3·6) (three patients)9·9 (8·6–16·6) (three patients)12·3 (10·1–26·3) (four patients)
Total body1·0 (1·0–1·1)1·7 (1·5–1·8)2·4 (2·2–2·5)
Figure 2.

 Haematological toxicity. White blood cell (A), neutrophil (B) and platelet (C) recovery for level of activity administered: 30 MBq/kg, 45 MBq/kg and 56 MBq/kg. Reinfusion was performed on days 0 and 13 after treatment.

Results

Patient characteristics

Between April 2004 and August 2005, 13 patients were enrolled in the study: four at the first, four at the second and five at the third activity level. Their characteristics were well balanced between the three levels and are reported in Table I. Median age was 68 years (range 28–73 years), 85% were male. All patients, except one, had advanced stage disease (III/IV) at diagnosis and a median number of 3 (range 2–6) prior therapies had been administered. Histology included: one FL, eight DLBCL, three mantle-cell lymphoma (MCL) and one transformed-marginal lymphoma. Median time from diagnosis to study entry was 28 months (range 9–49). Three patients had received prior irradiation and six had received HD-CT with ASCT. All the patients had received prior Rituximab either alone or in combination with CT. BM biopsies were negative for disease localization in all patients before RIT.

At time of treatment two patients had ANC < 1·0 × 109/l (both at the third dose level), eight patients had PLT between 100 and 150 × 109/l (two at the first level, three at the second and three at the last dose level) and one patient had Hb < 100 g/l (at the second dose level).

Among the 13 patients treated, eight were mobilized and stem cells were harvested before 90Y-ibritumomab-tiuxetan administration (seven had never received a transplantation before, while one underwent a second mobilization course and leukapheresis). The remaining five patients had cryopreserved stem cells from the previous mobilization and autologous transplantation. Three patients were not included in the trial because of a mobilization failure.

Dosimetry

Dosimetry using 111In-ibritumomab tiuxetan was performed in all patients, showing favourable radiation-absorbed doses to uninvolved organs in all cases. However, one patient, planned for second level dose administration, and one patient planned for third level dose administration, were treated at a lower activity level because of elevated liver captation. The median adsorbed doses and ranges of variability for 90Y-ibritumomab tiuxetan were (mGy/MBq): 0·90 (0·50–0·95), red marrow; 2·4 (1·6–5·4), heart wall; 1·7 (0·3–2·7), lungs; 3·3 (2·0–10·6), liver; 2·2 (1·0–5·0), spleen; 1·5 (0·6–3·7), kidneys; 2·8 (1·3–4·7), testes; 0·5 (0·4–0·8), total body (TB). Table II shows radiation-absorbed doses of uninvolved organs as per activity level. Interestingly, only one patient reached the dose of 30 Gy to the liver without developing toxicity.

Four patients received an activity of 30 MBq/kg, four patients 45 MBq/kg and five patients 56 MBq/kg. The median activity of 90Y-ibritumomab-tiuxetan delivered was 3·3 GBq, range 2·1–5·55 GBq, as four patients received an activity below 2·96 GBq, four within 2·96–3·7 GBq and five >3·7 GBq.

Engraftment and haematological toxicity

A median of 5·25 × 106/kg (range 1·94–20 × 106/kg) CD34+ cells were infused for each patient. Interestingly, we did not notice any substantial differences in engraftment between those patients receiving <3 vs. >5 × 106/kg CD34+ and there was no relationship with the 90Y-ibritumomab-tiuxetan activity received.

Median time to engraftment was 12 (range 0–22) days and 19 (range 0–48) days for PLT and ANC respectively. A delay occurred in only one patient, who received 56 MBq/kg with ANC at baseline of 0·20 × 109/l: she reached ANC > 1·0 × 109/l 48 d after reinfusion. PLT and ANC-count nadirs were reached 8 and 4 d after transplantation, with median values of 11 × 109/l PLT (range 4–35) and 0·01 × 109/l ANC (range 0·01–1·09).

When data were analysed per dose level, low numbers of occurring events did not enable a complete statistical analysis, however no clear differences in terms of haematological toxicity were observed (Fig 2A–C, Table IIIA and B).

Figure 2A and B presents detailed information on duration of leucopenia–neutropenia in relation to the activity administered. All patient, excepted one enrolled at the first dose level, experienced at least a grade 3 ANC toxicity (<1·0 × 109/l). The median duration of grade 3–4 neutropenia was 15 d starting from day +1 from transplantation (median value). Some patients experienced a late drop in WBC and ANC counts. However, they did not develop infections or require GCSF treatment, showing a slow, but spontaneous recovery to normal values.

Figure 2C shows the curves of PLT toxicity as per activity level. Patients receiving 30 MBq/kg reached PLT > 100 × 109/l with a median of 26 d (range 19–29), those receiving 45 MBq/kg with a median of 19 d (range 12–37) and those receiving 56 MBq/kg with a median of 21·5 d from reinfusion (range 16–152).

Interestingly, PLT count at baseline seemed to have an important role on developing toxicity. In fact, all patients who presented a normal PLT count at baseline recovered quickly (range 16–32 d), without any difference between activity-levels. On the contrary, eight patients who presented a PLT count below 150 × 109/l at baseline, had a slower recovery and never reached a value >150 × 109/l PLT, except one receiving 30 MBq/kg (data not shown). However, two patients treated at the second and third dose level, never reached PLT > 100 × 109/l, because they progressed at 35 and 139 d after 90Y-ibritumomab tiuxetan respectively, at a PLT of 75 × 109/l (PLT at baseline was 100 and 120 × 109/l respectively).

A drop in PLT often occurs after engraftment as a sign of possible late toxicity, particularly in those patients who received more than three previous CT regimens (Figs 2C and 3). Five patients required red blood cell transfusions: one patient at the first level, two at the second and two at the third dose-level, with a median of 4 each (range 2–4). Eight patients required PLT transfusions (two at the first level, two at the second and four at the third activity level) with a median of 1 each (range 0–4).

Figure 3.

 Platelet count recovery at 56 Mbq/kg activity level and number of previous computed tomography scan regimens.

Non-haematological toxicity

During 90Y-ibritumomab-tiuxetan administration, none of the patients experienced infusion-related toxicity and was forced to discontinue the treatment. In relation to acute non-haematological toxicity, infections and liver toxicity were the most relevant. In particular, one febrile neutropenia and one herpes zoster virus (HZV)-infection occurred at the second level, one febrile neutropenia at the third level. Acute liver toxicity was mild (grade 1) in one patient at the first and in one patient at the second level (both of them received <11 Gy to the liver) and was severe (grade 3) in one patient at the third level who received 15 Gy to the liver. However they all recovered quickly (Table IVA).

In relation to late non-haematological toxicity, one patient at the third level developed bacterial pneumonia followed by HZV-infection, two and 5 months after 90Y-ibritumomab tiuxetan respectively (Table IVB). One patient with DLBCL who received 56 MBq/kg (5·55 GBq total activity) experienced a late grade 4 thrombocytopenia and liver impairment 50 d after transplantation. BM examination showed hypocellular marrow without lymphoma involvement or sign of myelodysplasia. Serology identified a reactivation of hepatitis C virus infection and no signs of veno-occlusive disease were detected. This patient died in CR, 4 months (124 d) after treatment (Table IIIB). However, he did not receive a higher-dose to the liver (17 Gy) with respect to the previously reported one. Dose received by the red marrow was 5 Gy.

Table IV.   (A) Acute non-haematological toxicity; (B) Late non-haematological toxicity.
 30 MBq/kg (= 4)45 MBq/kg (= 4)56 MBq/kg (= 5)
  1. FUO, fever of unknown origin; HZV, herpes zoster virus; HCV, hepatitis C virus.

(A)
Mucositis01 grade 11 grade 1 and 1 grade 2
Infections01 HZV0
1 HSVI
Febrile neutropenia011
FUO000
Liver1 grade 11 grade 11 grade 3
1 grade 1
Asthenia02 grade 10
Nausea01 grade 20
(B)
Infections001 HZV
1 bacterial pneumonia
1 reactivation HCV
FUO001
Table III.   Time to (A) platelet count (PLT) nadir. Log rank test = 0·9750; (B) absolute neutrophil count (ANC) nadir. Log rank test = 0·3829.
Dose levelTime to nadir (d)Statistics
n(Mean ± SD)Median (95% CI)PLT (mean ± SD)
  1. SD, standard deviation; 95% CI, 85% confidence interval.

(A)
30 MBq/kg420·0 ± 2·322·00 (13·0–23·0)16·7 ± 13·9
45 MBq/kg421·2 ± 1·221·05 (18·0–24·0)14·5 ± 3·4
56 MBq/kg532·0 ± 13·520·00 (15·0–86·0)9·8 ± 5·4
(B)
30 MBq/kg415·0 ± 1·515·00 (12·0–18·0)Undetectable
45 MBq/kg419·5 ± 2·217·05 (17·0–26·0)0·12 ± 0·24
56 MBq/kg516·6 ± 1·916·00 (12·0–21·0)0·04 ± 0·09

No pulmonary, gastro-intestinal, renal or cardiac severe toxicity has been observed so far. However, one patient developed a congestive heart failure 12 months after therapy. He was affected by DLBCL and had already received an ASCT before 90Y-ibritumomab tiuxetan. He experienced an early disease progression after RIT and underwent a new induction with a third ASCT. Congestive heart failure was secondary to the latter approach and contemporary to multi-organ failure and sepsis. The patient has recently, but not completely recovered from this.

Finally, one patient developed a myelodysplastic syndrome – refractory anaemia with excess of blasts – 2 years after treatment, when still in CR for MCL. Cytogenetic diagnosis showed a 46,X,t(X;20)(q13;q12),del(7)(q22)[22] karyotype. She had received previous HD-CT with ASCT and alkylating-containing regimens. The total activity she received was 45 MBq/kg, while the dose to the red marrow was <2 Gy.

HAMA development

All patients had serial testing for HAMA but none had developed antibodies to date.

Response

All patients were evaluable for response after a median follow-up of 9 months (range 2–24 months). Eight of 13 patients experienced a clinical and radiological response with a 61·5% overall rate. In particular, six patients obtained a CR and two patients a PR. Two of those six patients who reached a CR remained DF 30 months from 90Y-ibritumomab tiuxetan, one died in CR at 4 months for a reactivation of a known hepatitis, while the other three relapsed at 6, 14 and 18 months after treatment. The two patients who only reached a PR progressed 4 and 9 months after treatment.

When data were analysed per activity level we observed: two CR and two PD at the first level; two CR and two PD at the second level and two CR, two PR and one PD at the third level.

Despite the limited number of patients, efficacy seemed to be higher in MCL, independently of the administered activity. In fact, two of three patients affected by MCL obtained a CR, one of whom is still alive and DF after 2 years, while the second one progressed 14 months after treatment. This data support previously published experiences with Bexxar® (GlaxoSmithKline, Philadelphia, PA, USA) which were focused on MCL and demonstrated longer OS and DF survival (Gopal et al, 2002; Dreyling et al, 2005).

Discussion

Early studies pioneered the use of HD 131I-labelled-tositumomab (Bexxar®), strongly suggesting the existence of a direct dose/response curve when RIT was administered at higher activities in respect to standard ones (Press et al, 1993, 1995; Liu et al, 1998).

A study of the myeloablative activities of 90Y-ibritumomab tiuxetan as a single agent has not been reported to date; here we provide the first evidence of its safety profile and potential clinical efficacy as an ASCT-conditioning regimen for patients with relapsed or refractory NHL.

Our results showed that, even at 4-times higher activity than the standard, high activities of 90Y-ibritumomab tiuxetan can be safely delivered to patients with a severe prognosis, including those (elderly and heavily pretreated ones) considered unsuitable to receive HD-CT with ASCT (Gulati et al, 1992; Freedman et al, 1999; Shipp et al, 1999; Winter, 2004).

Three escalating 90Y-ibritumomab-tiuxetan activity levels were chosen based on previously reported data using standard-activity treatment (14·8 MBq/kg) (Wiseman et al, 2003) and RIT/CT combination approaches (Winter et al, 2001, 2004; Devizzi et al, 2005; Krishnan et al, 2005; Nademanee et al, 2005; Shimoni et al, 2005). Our experience in treating solid tumours with Yttrium-based regimens was also taken into consideration (Paganelli et al, 2006).

Reinfusion was performed 13 d after treatment, to prevent haematological toxicity, avoiding early cell killing by persistent circulating radioactivity (Shen et al, 2005). Growth factors were not suggested to avoid misunderstanding toxicity. However, haematological toxicity was completely and rapidly overcome by stem-cell reinfusion, showing no significant delayed engraftment between the three groups. This indicated that circulating radioactivity, as confirmed by dosimetry, does not influence reinfused stem-cell viability.

Dosimetry was performed in all patients and was helpful in confirming the planned profile of activity to be administered. If the non-target organ uptake was considered unfavourable, the patient could not be treated any more at that dose, could still be treated at decreased activity (data not shown). The limits for the irradiation of the liver, as well as of other vital organs (red marrow, kidneys, lungs, spleen and testes), were chosen as a conservative choice to prevent unexplored risks and were evaluated patient by patient. In fact, irradiation of vital organs in patients undergoing HA-RIT, deserves special caution because of the clinical history of the patients who received many lines of previous chemotherapy. In particular, dosimetry studies give important information on the suggested dose for the red marrow to ensure an adequate engraftment: should the dose be >5 Gy, recovery, particularly PLT, could be greatly delayed (Juweid, 2002).

Interestingly, the time of PLT recovery and transfusion requirements also seemed to be influenced by the number of previous CT-regimens administered, as a sign of progressive BM impairment. Non-haematological toxicity was manageable, including infections and hepatic toxicity; the latter should be probably considered the new dose-limiting rationale.

In fact, our dosimetry evaluations demonstrated that higher-RIT doses could enhance non-specific hepatic uptake, causing, at least, a transiently impaired function or a strong reactivation of hepatitis infection previously detected as a chronic disease.

In our opinion, three points emerged from this study: (i) haematological toxicity could remain a problem in those heavily pretreated patients (more than three CT lines) enrolled when the PLT is under 150 × 109/l; (ii) liver is the second dose-limiting organ especially when affected by chronic diseases, such as hepatitis B or C; and (iii) dosimetry should always be performed to avoid unpredictable toxicity.

Based on these data, we suggest that the full dose of 56 MBq/kg should be administered only to those patients who have received no more than three previous lines of treatment. Possibly, they should have a baseline PLT ≥ 150 × 109/l. The dose of 45 MBq/kg could be reserved for patients who have received more than three lines of therapy and/or have PLT between 100 and 150 × 109/l. We believe that it is not advisable to administer elevated Zevalin® activity to patients with a PLT < 100 × 109/l and/or affected by chronic diseases of the liver.

Although assessment of efficacy was not a primary objective of this study and the number of patients is too small, preliminary results seem encouraging, As six of 13 patients achieved a CR; for two of these patients the CR has lasted more than 2 years after treatment. In particular, among the five evaluable patients who received the higher-activities, two achieved a CR and two a very good PR, suggesting the possible relationship between higher activity and best response. In fact, when evaluating these results, it must be considered that the majority of patients enrolled in the trial had no real other therapeutic option to be offered.

Moreover, we have to underline that high activities of 90Y-ibritumomab-tiuxetan does not preclude patients from receiving further therapies including highly myelotoxic ones, demonstrated for standard Y-based therapies (Ansell et al, 2002).

In conclusion, treatment-related toxicities and clinical activity suggest HA 90Y-ibritumomab tiuxetan to be an interesting modality treatment for further investigation as a different therapeutic option for resistant/refractory NHL considered unsuitable for aggressive salvage treatments. We are currently expanding accrual at the higher level to confirm the toxicity profile and evaluate the efficacy on a larger cohort of patients.

Acknowledgements

Authors wish to thank Liliana Calabrese for data management and Sarah Liptrott for critical reading of the manuscript and English control. We are grateful to ‘Comitato Grazia Focacci per il Rotaract’, ‘Umberto Veronesi Foundation’ and Bayer Schering Pharma for their support.

Ancillary