Clinical Trial of Consolidation Treatment With Iodine I 131 Tositumomab for Multiple Myeloma

Multiple myeloma is ranked second in hematological malignancies in the United States (Munshi
et al., 2001). Because no curable option exists, patients with early stage disease are
typically observed without treatment until the disease progresses or symptoms appear.

The majority of patients will respond to initial chemotherapy, however, all treated patients
repeatedly relapse with shorter remissions following each subsequent course of chemotherapy
(Rajkumar et al., 2002a). Regimens most commonly used for initial therapy include melphalan
and prednisone (MP), which induce remission in approximately 50% of patients (Alexanian et
al., 1969), VAD—vincristine, adriamycin and dexamethasone with response rates in the 50%
range (Salmon et al., 1994; Barlogie et al., 1984; Alexanian et al., 1986), dexamethasone
pulses, and more recently introduced combinations of thalidomide with dexamethasone (Rajkumar
et al., 2002b; Weber et al., 2003). All these regimens have comparable response rates. With
the exception of the recent report, which showed a superior response rate to the combination
of thalidomide with dexamethasone versus dexamethasone pulses (Rajkumar et al., 2004),
multiple other comparative studies did not show superiority of any of these regimens for
overall survival. Complete responses to any of these regimens are observed infrequently (<
5%). Patients who complete initial therapy, usually proceed to high dose chemotherapy with
autologous stem cell transplant support. Alternatively, patients may be placed on maintenance
regimen or can be observed. Except for one study with alternate day high dose of Prednisone
(Berenson et al., 2002), no convincing evidence exists that maintenance impacts the natural
history of the disease (Kyle, 2002; The Myeloma Trialists' Collaborative Group, 2001).

High dose chemotherapy has been used as consolidation in multiple myeloma by a number of
investigators (Barlogie et al., 1986; Cunningham et al., 1994; Harousseau et al., 1995). A
randomized trial conducted by French Myeloma Intergroup (IFM 90) provided evidence that
treatment with high dose chemotherapy and autologous stem cell transplant (ASCT) after
initial therapy results in significantly longer event-free survival (median 28 vs. 18 months)
and overall survival (57 vs. 42 months) if compared to conventional chemotherapy (Attal et
al., 1996). The benefit of auto transplantation was seen in patients of all ages (Siegel et
al., 1999). More recently, similar results were reported from a large randomized study in
Great Britain (Childs et al., 2003). Based on these studies, in patients eligible for
transplant, consolidation of initial therapy with high dose therapy followed by ASCT is
considered a standard of care (NCCN Guidelines 2004). However, ASCT is not curative even in
patients who achieve complete response (CR), which is observed in 20-30% of patients who
completed ASCT. Contamination of autologous stem cell collection with malignant myeloma cells
may play some role. In support of this notion, syngeneic bone marrow transplants showed
better and more durable outcomes than those seen with ASCT (Gahrton et al., 1999). Another
reason for failure of autologous stem cell transplant is ineffective eradication of myeloma
clone with high dose chemotherapy. In an effort to provide more effective elimination of
malignant clones, some investigators developed intensified treatment regimens. Improved
results have been reported using Total Therapy, which involves a sequence of treatments,
including tandem ASCT (Barlogie et al., 1999, Barlogie at al., 2004). The notion that
intensification of therapy improves outcomes has been validated in a recent randomized study,
which showed that consolidation of initial therapy with tandem autologous stem cell
transplant was superior than consolidation with single autologous stem cell transplant as
measured by an improved duration of responses and overall survival (Attal at al, 2003). In
contrast to ASCT, allogeneic stem cell transplantation can provide stem cell products free of
malignant cells, as well as the benefit of graft-versus-myeloma effect.

However, very high 1-year mortality (30-60%) was reported in early allogeneic bone marrow
transplant studies, with disease free survival in only 15-30% of patients and no superior
survival if compared to ASCT (Bensinger et al., 1996; Bjorkstrand et al., 1996; Barlogie et
al., 1995). More recent studies, including allogeneic transplants with nonmyeloablative
conditioning regimens, provide more encouraging results (Reynolds et al., 2001; Badros et
al., 2002). Recently, a combined treatment with ASCT followed by a non-myeloablative
allogeneic stem cell transplant showed very promising outcomes and improved, although still
significant toxicity, with CR rates in 55-60% range (Kroger et al., 2002; Maloney et al.,
2003). It is too early to determine whether improved responses using this approach will
result in an increased survival.

Emerging data suggest that achieving CR or near CR (nCR) will result in more durable
remission and longer survival. Recent analysis of long term outcomes of patients treated on
randomized French study IMG90 clearly indicates that the longest survivors are those patients
who achieve at least very good partial response or complete response (Harousseau, 2003). At 7
years, 60% of patients were alive in a group, which achieved CR or near CR. In contrast, in a
group of patients who achieved > PR but less than CR or near CR only 30% were alive and among
patients who had less than PR none were alive. Similar results have bee recently reported for
patients who underwent allogeneic stem cell transplant (Corradini et al., 2003). It is not
clear whether CR in response to initial therapy and prior to transplant may have similar
impact on overall outcomes. Regardless of specific transplant approach, survival curves do
not plateau and all patients are expected to relapse, regardless of method of intensification
therapy, possibly due to persistence of residual population of chemotherapy-resistant
clonogenic myeloma cells.

For patients who have relapsed or are refractory to therapy, there is no agreed upon standard
treatment (Anderson et al., 2002; Kyle, 2002). Treatment options include salvage
chemotherapy, autologous stem cell transplant (if not previously done or as a second
transplant), or allogeneic stem cell transplant, full or low intensity (Kyle, 2002). Salvage
chemotherapy is most widely used in clinical practice. Among a variety of salvage regimens,
both monotherapy and combination therapy have been applied. Monotherapy with Dexamethasone or
other steroids administered as pulse therapy produced responses in the 35-40% range
(Alexanian et al., 1983; Gertz et al., 1995). Thalidomide used as a single agent showed a 32%
response rate in this patient population (Singhal et al., 1999). More recently, VELCADE as a
single agent induced at least minimal responses (i.e. > 25% reduction in monoclonal protein)
in 35% of patients and at least a stabilization of the disease in 59% of patients with
relapsed/refractory multiple myeloma using strict SWOG criteria (Richardson et al., 2003).
Combination therapies historically show higher response rates. VAD has been demonstrated as
an effective regimen in patients refractory to alkylating agents, with response rates of 60%
(Lokhorst et al., 1989). A similar regimen called DVd, with Doxil, Vincristine, and
dexamethasone, showed comparable efficacy and acceptable toxicity (Hussein et al., 2002;
Rifkin et al., 2004). Newer combinations, including combinations of VELCADE, thalidomide, and
Revlimid (analog of thalidomide) are promising, and appear to be able to induce higher
response rates and complete remissions as per unpublished yet reports from different meetings
(Agarwal et al, 2003; Orlowski et al, 2003, Richardson et al, 2003; Richardson et al., 2004).

Clonogenic Multiple Myeloma Cells:

Myeloma is characterized by an accumulation of malignant plasma cells in bone marrow.
Numerous observations indicated that malignant plasma cells have low proliferative capacity
and it is believed that the vast majority of the malignant cell population in myeloma is
represented by terminally differentiated plasma cells, similar to their normal counterparts
(Barlogie et al., 1989). It is unclear whether these terminally differentiated malignant
plasma cells are capable of self-renewal. During the past decade, several studies indicated
that multiple myeloma, similarly to other malignancies, may consist of heterogenous
population of malignant cells (Bakkus et al., 1994; Billadeau et al., 1996).

Clonotypic studies identified a population of circulating B-cells in blood samples of
patients with myeloma (Bergsagel et al., 1995, Chen and Epstein, 1996), sharing the same
clonotypic CDR3 region as is detected in the bone marrow malignant plasma cells (Pilarski et
al, 1996, Szczepek et al., 1998). Further studies provided additional insights into the
evolution of the myeloma clone supporting a notion of heterogeneity of the population of
clonal cells in myeloma (Taylor et al., 2002). The B-cell component of the myeloma clone
appeared to be clonogenic in xenografted mice (Pilarski et al., 2000; Reiman et al., 2001).
Moreover, clonotypic cells with B-cell phenotype express CD19+/CD20+ antigens and may
represent a reservoir of disease that persists after therapy, including high dose
chemotherapy (Kiel et al., 1999; Rottenburger et al., 1999; Pilarski et al., 2002).

The majority of myeloma cells express CD138, a highly specific surface antigen of terminally
differentiated plasma cells, which is absent on highly proliferative normal plasmablast and
earlier stages of B-cells (Jego et al., 2001). Based on these observations, it was
hypothesized that clonogenic myeloma cells should lack CD138 expression. Using CD138+ and
CD138- subsets of cells isolated from myeloma cell lines and from primary myeloma patient
samples, Matsui et al. (2004), showed that only a minority of all myeloma cells have CD138-
phenotype (<5%). However, only CD138- have clonogenic potential, as demonstrated in colony
formation assays in methylcellulose and after transplantation to NOD/SCID mice. Moreover,
these cells express B-cell antigens, including CD19 and CD20 and their growth could be
inhibited by in vitro treatment with rituximab, an antibody directed against CD20 antigen
(Matsui et al., 2004).

Rationale for anti-CD20 Therapy in Myeloma:

Recent observations that clonogenic cells in myeloma express CD20 (Kiel et al, 1999,
Rottenburger et al, 1999) and that their growth could be inhibited in vitro by rituximab, an
anti-CD20 antibody (Matsui et al, 2004), provides a rationale for using CD20-directed therapy
in patients with myeloma. Moreover, CD19+/CD20+ clonotypic cells appear to be more refractory
than CD19-/CD20- cells to commonly used chemotherapy regimens in myeloma including high dose
therapy with stem cell transplant (Kiel et al., 1999, Rottenburger et al, 1999). Therefore,
CD20+ directed treatment of myeloma would complement chemotherapy by targeting chemoresistant
clonogenic myeloma cells.

Previous studies showed that therapy with Rituximab, unlabeled anti-CD20 antibody, could be
active in multiple myeloma (Hussein et al. 1999, Treon et al, 2002).

In this study, we propose to apply therapy with Bexxar, a radiolabeled anti-CD20 antibody.
Bexxar Therapeutic Regimen is composed of the murine anti-CD20 monoclonal antibody
Tositumomab and Iodine I 131 Tositumomab. Iodine I 131 Tositumomab is a radio-iodinated
derivative of Tositumomab that has been covalently linked to Iodine-131.

We hypothesized that radiolabeled anti-CD20 antibody will be more efficacious than unlabeled
antibody (i.e. rituximab) in eradication of highly radiosensitive myeloma cells, similarly to
what was observed in non-Hodgkin's lymphoma (Horning et al., 2000; Davis et al, 2001 Flinn et
al, 2001; Witzig et al, 2002). It is presumed, that higher efficacy of radiolabeled anti-CD20
antibody, if compared to unlabeled antibody, may be due to "crossfire effect" as the
radiation is delivered not only to target cells but also to neighboring cells, both
expressing and not-expressing CD20 (Vose et al, 1999). In the proposed design, we will apply
Bexxar after completion of a course of initial chemotherapy for newly diagnosed patients or a
course of salvage chemotherapy, provided that patients achieve at least partial response to
prior therapy (i.e. > 50% reduction of tumor mass) and reach a plateau of their response to
prior therapy. By using patients in a period of stable disease, we will have a better chance
to observe the effects of radioimmunotherapy against CD20+ cells, which represent a minority
of all malignant cells. In addition, we will evaluate the impact of Bexxar therapy on
clonogenic myeloma cells using clonogenic assays as described in Methods.

We anticipate that majority of newly diagnosed patients will be eligible and will
subsequently proceed to autologous stem cell transplant as part of standard therapy of
myeloma (NCCN Guidelines 2004) if, as hypothesized, bexxar treatment does have the potential
to eliminate clonogenic myeloma stem cells; stem cells collected post-bexxar treatment should
be depleted of re-populating myeloma cells. Therefore, this combination of chemotherapy and
targeted therapy against clonogenic cells may have a potential of eliminating myeloma clones.
Although stem cell collections and autologous stem cell transplants have been used without
apparent difficulties in the past in patients treated with anti-CD20 radio-conjugates
(Ratanatharatorn et al, 2001; Kaminski et al, 2001; Ansell et al, 2002), we plan to collect
backup stem cells prior to treatment with Bexxar, to allow to proceed with the standard stem
cell transplant in the unlikely cases of failure of collection of stem cells post-Bexxar.
Anti-CD20 radioimmunotherapy was previously used in combination with chemotherapy without
additional toxicity (Emmanoulides et al, 2001; Gregory et al, 2001, Press et al, 2003).

Radioimmunotherapy of B-Cell Malignancies:

Radiolabeled monoclonal antibodies are efficacious in treating B-cell malignancies for the
following reasons: B-lymphocytes, lymphoma, and myeloma cells are inherently sensitive to
radiotherapy (Parker et al, 1980); the local emission of ionizing radiation by radiolabeled
antibodies may kill cells with or without the target antigen in close proximity to the bound
antibody; and penetrating radiation may obviate the problem of limited access in bulky or
poorly vascularized tumors.

Early investigations of therapeutic radiolabeled antibodies for B-cell lymphoma were
performed with iodinated antibodies (Press et al, 1989; Kaminski et al, 1993; Wahl et al,
1994; Press et al, 1995). Currently, two treatment regimens have been approved for treatment
of relapsed/refractory and transformed follicular lymphoma. The Bexxar Treatment Regimen
includes murine anti-CD20 antibody Tositumomab iodinated with Iodine I 131. The Main
anti-tumor effect comes from high energy beta particles emitted by I 131. In addition, I 131
is emitting also low energy gamma radiation, allowing for gamma camera measurements and
calculation of individualized dose of radioactive tracer for a given patient during
therapeutic phase (see below). In Zevalin, a murine anti-CD20 antibody ibritumomab is
covalently linked to 90Yttrium (90Y), which is a pure beta emitter. Lack of gamma radiation
from 90Y does not allow for dosimetric assessments and a fixed dose of Zevalin is used for
all patients based on their weight. This can possibly result in either under- or overdosing
of some patients.

In addition to treatment of follicular lymphoma, ongoing studies explore a possibility of
treatment with either Bexxar or Zevalin of other B-cell malignancies. In particular,
anti-CD20 radioimmunotherapy appears very promising as part of treatment of mantle cell
lymphoma, large cell lymphoma, and Waldenstrom's macroglobulinemia.

Iodine I 131 Tositumomab

Background:

Tositumomab is a murine IgG2a lambda monoclonal antibody directed against the CD20 antigen,
which is found on the surface of normal and malignant B lymphocytes. Tositumomab is produced
in an antibiotic-free culture of mammalian cells and is composed of two murine gamma 2a heavy
chains of 451 amino acids each and two lambda light chains of 220 amino acids each. The
approximate molecular weight of Tositumomab is 150 kD. In vitro studies have demonstrated
that upon binding to the CD20 antigen, Tositumomab is capable of inducing apoptosis. In
addition, Tositumomab induces antibody-dependent cellular cytotoxicity (ADCC) and
complement-dependent cellular cytotoxicity (CDC). Iodine I 131 Tositumomab is a
radio-iodinated derivative of Tositumomab that has been covalently linked to Iodine-131.
Unbound radio-Iodine and other reactants have been removed by chromatographic purification
steps. This agent produces its cytotoxic effect through the antitumor effects of ionizing
radiation, as well as the more direct antitumor effects of the antibody. The goal of
treatment with Iodine I 131 Tositumomab is selective delivery of radiotherapy to
radiosensitive malignant cells, thus minimizing toxicity to the normal organs.

Bexxar Clinical Experience:

The dosing regimen and maximally tolerated dose of Bexxar were established in a phase I/II
single-center study conducted at the University of Michigan (Kaminski et al, 2000). The
RIT-II-001 trial included 47 patients and was designed to validate the dosing methodology
developed at Michigan (Vose et al, 2000). The RIT-II-002 trial randomized 78 patients to
receive either the tositumomab and I 131-tositumomab regimen or the unlabeled tositumomab to
determine the value added of the radionuclide component (Davis et al, 2001). Patients treated
on the radiolabeled antibody arm showed a higher overall response rate (OR = 55% vs. 19%) and
complete response rate (CR = 33% vs. 8%). The RIT-II-004 study enrolled 60 patients with
chemotherapy-refractory disease (no response or response lasting less than 6 months to the
last chemotherapy received) to assess the efficacy of the Bexxar therapeutic regimen in this
patient population (Kaminski et al, 2001). The number of patients who achieved longer
duration of response to Bexxar was about 5 times higher than the number of patients who had a
longer duration to chemotherapy (p <0.001). In addition, patients treated with Bexxar
achieved a higher overall response (OR) rate (47% vs. 12%) and complete response (CR) rate
(20% vs. 2%) than patients treated with chemotherapy. In summary, all 4 initial studies,
including the Phase I/II trial, showed high response rates and duration of responses in
patients with relapsed or refractory low grade or follicular lymphoma, including transformed
follicular lymphoma previously-treated with chemotherapy. Most remarkably, patients who
achieved complete responses experienced often particularly long duration of responses lasting
for years. In recognition of remarkable activity, the FDA approved an expanded indication for
the Bexxar Therapeutic Regimen on January 3, 2005.

Inclusion Criteria:

- Age ≥ 18 years

- Expected survival ≥ 6 months

- Pre-study performance status of 0, 1, or 2 according to the World Health Organization
(WHO)

- Newly diagnosed or relapsed/refractory myeloma with histologic confirmation of
multiple myeloma by the Department of Pathology at University of Michigan Cancer
Center (UMCC)

- Not more than 3 lines of therapy for myeloma for patients with relapsed disease

- Documented Stage II or III multiple myeloma (Durie and Salmon, 1975) prior to
initiation of first line therapy

- At least 4 cycles of first line (for newly diagnosed patients) or salvage (for
relapsed/refractory patients) prior therapy and in a plateau of at least partial
response (Blade et al, 1999) for at least 2 determinations 6 weeks apart

- At least 21 days from day 1 of the last cycle and fully recovered from all toxicities
associated with prior surgery, radiation treatments, chemotherapy, or immunotherapy

- Measurable M-proteins with greater than 1 g/dl serum monoclonal protein and/or greater
than 0.5 g/24 hour urine light chain excretion

- Acceptable hematologic status within two weeks prior to patient registration,
including:

- Absolute neutrophil count ([segmented neutrophils + bands] x total white blood
cell [WBC]) ≥ 1,500/mm3;

- Platelet counts ≥ 150,000/mm3; these patients will receive total body dose of 75
cGy of Bexxar; or

- Platelet counts from 100,000/mm3 to 149,000/mm3; these patients will receive a 65
cGy total body dose of Bexxar;

- In patients previously treated with ASCT, total body dose will be 55 cGy in
patients with platelet count > 150,000 and 45 cGy in patients with platelets
100,000-149,000.

- Female patients who are not pregnant or lactating

- Men and women of reproductive potential who are following accepted birth control
methods (as determined by the treating physician)

- Patients previously on Phase II drugs if no long-term toxicity is expected, and the
patient has been off the drug for three or more weeks with no significant post
treatment toxicities observed

- Patients determined to have < 25% bone marrow involvement with myeloma within six
weeks of registration (based on bilateral core biopsy).

Exclusion Criteria:

- Patients with impaired bone marrow reserve, as indicated by one or more of the
following:

- Platelet count < 100,000 cells/mm3;

- Hypocellular bone marrow;

- Marked reduction in bone marrow precursors of one or more cell lines
(granulocytic, megakaryocytic, erythroid);

- History of failed stem cell collection;

- Myelodysplastic syndrome (MDS) or evidence of other than myeloma clonogenic
abnormalities;

- Prior radioimmunotherapy;

- Prior anti-CD20 therapy;

- Other than myeloma malignancy, except B-cell non-Hodgkin's lymphoma, basal and
squamous cell carcinoma of the skin, and cervical and breast cancer in situ,
unless patient is cancer free for > 3 years;

- Central nervous system (CNS) involvement;

- Patients with known HIV infection;

- Patients with pleural effusion;

- Patients with abnormal liver function: total bilirubin > 2.0 mg/dL;

- Patients with abnormal renal function: serum creatinine > 2.0 mg/dL;

- Patients who have received prior external beam radiation therapy to > 25% of
active bone marrow (involved field or regional);

- Patients who have received G-CSF or GM-CSF therapy within two weeks prior to
treatment;

- Serious nonmalignant disease or infection which, in the opinion of the
investigator and/or the sponsor, would compromise other protocol objectives;

- Major surgery, other than diagnostic surgery, within four weeks;

- Presence of anti-murine antibody (HAMA) reactivity. This result must be available
prior to receiving treatment for those patients with prior exposure to murine
antibodies or proteins.