Open Label Study of Single Agent Oral RG7388 in Patients With Polycythemia Vera and Essential Thrombocythemia

The Philadelphia chromosome-negative chronic myeloproliferative neoplasms (MPNs) are a group
of hematopoietic stem cell malignancies that include polycythemia vera (PV), essential
thrombocythemia (ET), and primary myelofibrosis (PMF). PV and ET can evolve into
myelofibrosis, termed post PV/ET MF. ET, PV and PMF have variable tendencies to transform to
blast phase disease with a dismal prognosis. JAK2V617F is a point activating mutation
resulting in the constitutive activity of the JAK-STAT pathway within hematopoietic cells in
approximately 96%, 50%, and 50% of patients with PV, ET, and MF, respectively.

Polycythemia Vera is characterized by an absolute increase in red cell mass. Patients with PV
have a median survival if untreated of approximately 18 months from the time of diagnosis and
treated of approximately 18 years. PV-related symptoms include headache, weakness, dizziness,
epigastric distress, and pruritus. PV-related signs include hypertension, gout, left upper
abdominal quadrant pain, high hematocrit, leukocytosis, and thrombocytosis. Major causes of
reduced survival include thrombosis (29%), bleeding (7%), evolution to myelofibrosis (3%),
transformation to acute leukemia (23%), and solid tumors (16%). PV patients are stratified
for risk of thrombosis by age >60 and history of prior thrombotic events. Therapy for low
risk PV includes low dose aspirin and therapeutic phlebotomy to maintain a hematocrit <45% in
a man and 42% in a woman. Cardiovascular risk factor modification such as weight loss,
control of hypertension and hypercholesterolemia, and smoking cessation are also important
adjunctive approaches to all patients with PV. High risk patients are also treated with
cytoreductive therapy in the form of hydroxyurea to further reduce the risk of thrombotic
complications.

Essential Thrombocythemia is characterized by persistent isolated thrombocytosis and tendency
for arterial and venous thrombosis. A similar pattern of symptoms as noted above with PV are
also seen in patients with ET. The median survival of patients with ET is similar to that of
age and sex matched cohort and in some patients is limited by thrombotic complications (22%),
evolution to MF (10%), and acute leukemia (2%) (Barbui. J Clin Oncol. 2011; 29(23):3179).
Risk stratification for occurrence of thrombosis is based on age >60 years and/or history of
thrombosis. Additionally, cardiovascular risk factors and persistent thrombocytosis >1.5 x
109/L are believed to influence thrombotic risk and leukemic transformation has been shown to
be associated with anemia, older age, and leukocytosis. JAK2V617F is present in approximately
50% of cases and helps establish a diagnosis of a clonal thrombocytosis, and has been shown
in some studies to predict for a higher risk of thrombosis and potential for transformation
to PV. Management is aimed at reducing thrombotic risk with the use of low dose aspirin in
low risk patients (no risk factors) and cytoreductive therapy in high risk patients (at least
one risk factor). Hydroxyurea, anagrelide, and interferon (IFN) have all been used to
maintain a platelet count below 400 x 109/L in patients with a history of thrombosis
(secondary prophylaxis). Currently, hydroxyurea is considered standard of care for high risk
ET patients based on the results of the PT-1 study which demonstrated superiority of
hydroxyurea over anagrelide in arterial thrombosis risk reduction and worsening marrow
reticulin fibrosis in patients receiving anagrelide (Harrison N Engl J Med. 2005;353(1):33).
Hydroxyurea is associated with a risk of oral and skin ulcers, rash, and unacceptable
myelosuppression that can sometimes limit use in patients with ET/PV. Additionally, some
patients are unable to achieve adequate control of blood counts at doses below 2000 mg/daily
and this has been termed "resistance". Importantly, although a theoretical concern of
leukemogenic potential of hydroxyurea exists, based on the mechanism of action of this
chemotherapeutic agent, there are no definitive prospective studies clearly documenting an
increased risk of leukemic transformation.

More recently, a renewed interest in interferon-α for the treatment of PV as an alternative
therapeutic approach has led to the evaluation of pegylated interferon-α 2a (Pegasys, Roche)
in several phase II studies. Pegasys has an improved toxicity profile over intron-a and can
be self-administered by the patient on a weekly basis. Currently, Pegasys is being evaluated
in two large international trials within the myeloproliferative disorder research consortium
(MPD-RC). The MPD-RC 111 study is a phase II study intended to evaluate the response by
European LeukemiaNet (ELN) criteria in patients with high risk ET/PV who are intolerant or
resistant to hydroxyurea therapy treated with Pegasys. In addition, patients with documented
JAK2V617F and splanchnic vein thrombosis are also eligible for this clinical trial. MPD-RC
112 is a phase III study for newly diagnosed high risk PV/ET patients in which patients are
randomized to either Pegasys or hydroxyurea with a primary endpoint of response rate
comparison between the two treatment arms.

The use of intron-a (rIFN-α) and Pegasys has been extensively studied in patients with PV and
reported rates of discontinuation in the first year of therapy range from 14-40%. Objective
hematologic responses are seen in approximately 80% of treated patients and achievement in
complete phlebotomy free state in 60% of PV patients. Trials of Pegasys in the treatment of
PV have further demonstrated major molecular responses of 19% and complete eradication of
JAK2V617F in 14-24% of patients. Although hematologic remission can often be achieved within
months of starting rIFN-α treatment, molecular responses require longer term administration
and are rarely seen before completion of 12 months of therapy. Additionally, sustained
molecular remissions have been documented in patients that have discontinued therapy for up
to 30 months of follow up. In a retrospective review of 118 MPN patients receiving Pegasys
throughout multiple MPN centers included 55 PV patients with an ORR of 87% (54% CR, 33% PR)
by ELN criteria [21]. In this review, the most common non-hematologic toxicities were Grade
1-3 fatigue in 24 patients (20%), Grade 1 liver function test (LFT) elevation in 7 (6%), and
Grade 1-2 skin/allergic reaction in 6 (5%). Adverse effects leading to discontinuation were
primarily non-hematologic, although one patient (<1%) discontinued Pegasys therapy due to
Grade 2 anemia.

Although the biological mechanism of recombinant IFN in the treatment of PV is not completely
understood, evidence to suggest promotion of immunoregulatory cell function, inhibition of
angiogenesis, induction of pro-apoptotic gene expression, alteration of the bone marrow
microenvironment, suppression of hematopoietic progenitor cells, and increase cycling of
hematopoietic stem cells (HSC) has been reported.

In the last several years, oral small molecule tyrosine kinase inhibitors of JAK2 have been
tested in both pre-clinical and clinical trials of patients with MPNs including PV/ET. These
agents have demonstrated remarkable responses in terms of reduction in splenomegaly and
correction of leukocytosis, erythrocytosis, and thrombocytosis. However, mature follow up of
PV patients treated with JAK2 inhibitors evaluating reduction in thrombotic risk and
evolution to MF is not known. Currently, Ruxolitinib (Jakafi) is being evaluated in a phase
III study in patients with high risk PV that are refractory/intolerant to hydroxyurea
[Response trial: NCT01243944].

The tumor suppressor p53 plays an integral role in the regulation of the cell cycle,
apoptosis, DNA repair, and senescence. Fifty percent of cancers are found to have
inactivating mutations of p53. Although, inactivating mutations of p53 are uncommon in the
chronic phases of MPNs, they increase in frequency in the blast phase of MPNs. The p53
pathway can also be down-regulated in MPN cells that express wild-type p53 through
alternative pathways. Over-expression of the ubiquitin ligase murine double minute 2 (MDM2;
HDM2 in humans) is a regulator of wild type p53 through a variety of different mechanisms.
MDM2 1) promotes proteosomal degradation and 2) inhibits p53 transcription, and 3) inhibits
transactivation 4) facilitates export from the nucleus. Inhibition of MDM2 is expected to
release the negative regulation of p53 and ultimately promote tumor suppressor function. MDM2
overexpression can be a consequence of gene amplification, increased transcription, and
increased translation. It has previously been shown that in primary hematopoietic cells from
patients with JAK2V617F PV, reduced expression of p53 is a result of increased La autoantigen
expression leading to increased translation of MDM2. It would appear that JAK2V617F induces
this ribonucleoprotein and furthermore, La protein expression can be down-regulated by in
vitro JAK2 inhibitor treatment. Lu et al demonstrated increased MDM2 expression and reduced
p53 levels in JAK2V617F-positive PV CD34+ cells. The physical structure and interaction of
p53 and MDM2 is known and has allowed for the development of small molecule inhibitors of
MDM2-p53 interaction. Nutlin-3, a small molecule antagonist of MDM2, was shown to inhibit the
proliferation of PV CD34+ cells through increased p53 mediated apoptosis. Moreover, in vitro
combination of low doses of Pegasys with nutlin-3 resulted in selective and significant
inhibition of JAK2V67F-positive CD34+ PV colony formation compared to normal colony
formation. Thus, MDM2 appears to be a novel therapeutic target in JAK2V617F-positive PV and
ultimately may prove to be most effective when used in combination with Pegasys.

RG7388 is an MDM2-p53 binding cell cycle inhibitor (antiproliferative) and is expected to be
myelosuppressive at sufficiently high doses. Nonclinical safety findings with RG7388 that may
have potential clinical relevance include thrombocytopenia and neutropenia, as well as
general pancytopenia, and effects on liver function parameters.

Clinical experience to date suggests a dose relationship for RG7388 and gastrointestinal AEs
with increased incidence of nausea, vomiting, and diarrhea at higher dose levels;
gastrointestinal prophylaxis with anti-emetics is recommended. Cytopenias, manifesting as
thrombocytopenia and neutropenia, occurring at higher dose levels appear to be related to
exposure.

Iancu-Rubin et al have investigated the biological effect of RG7112 (first generation MDM2
inhibitor) induced p53-MDM2 disruption on megakaryopoiesis and platelet production in order
to better elucidate the mechanism of treatment associated thrombocytopenia (ASH 2012). Bone
marrow derived CD34+ cells exposed to RG7112 for 7 days generated fewer viable cells, fewer
CFU-MK colonies as compared to untreated cells. Based on these findings, it would appear that
RG7112 impacts megakaryopoiesis by two potential mechanisms: 1) Impairing the ability of
CD34+ cells to generate MK precursors due to increased apoptosis; 2) Limiting DNA synthesis
and polyploidization during the late stages of MK development due to pharmacological
activation of p53. A combination of these two effects may provide an explanation for
thrombocytopenia observed in patients receiving this drug and suggests that p53 plays an
important role in normal human thrombocytopoiesis.

Study NP27872 is a multi-center, open-label, first-in-human, Phase I dose-escalation study of
single agent RG7388, a small molecule MDM2 antagonist, administered orally in patients with
advanced malignancies except leukemia. The first patient received RG7388 on 15 November 2011.
As of 13 September 2012, 51 patients have been enrolled in the study.

Evaluable PK data is available for 22 patients in Study NP27872. There has been no evidence
of significant accumulation (defined as a mean ratio of area under the curve [AUC]τ [on Day
15 or 5] to AUCinf [on Day 1] > 2) observed in the study thus far, except for one patient who
exhibited a long half-life (t½) and consequently had an accumulated drug exposure that
increased 5-fold from Day 1 to Day 5. This was possibly due to concomitant controlled-release
oxycodone, as opioids are known to cause delayed gastric emptying and decreased intestinal
motility. In addition, data from the study demonstrate an approximately linear dose-exposure
(Day 1 maximum plasma concentration [Cmax]) relationship in RG7388 with moderately high
inter-patient variability without an absorption plateau to date.

Macrophage inhibitory cytokine (MIC-1, a secreted protein that is strongly induced by
activated p53) serum levels has been used to assess pharmacodynamic (PD) effects in this
Phase I study. Analysis of patients on 100 to 800 mg/day of RG7388 showed that the minimum
level for p53 induction occurs at a dose of 100 mg/day or a corresponding plasma level of 500
ng/mL of RG7388. Preliminary analysis of PK and safety data showed that there is an apparent
PK/PD relationship between an AUC per cycle and Cycle 1 platelet nadir.

At the data cutoff of 13 September 2012, 51 patients with advanced malignancies had received
R05503781 in Study NP27872. Fifty patients experienced at least 1 adverse event (AE) and 11
patients experienced at least 1 serious adverse event (SAE).

Thirteen SAEs in patients (white blood cell count decreased, platelet count decreased, 2
cases of neutrophil count decreased, febrile neutropenia, 5 cases of thrombocytopenia, 2
cases of neutropenia, and diarrhea) were considered by the investigator to be probably
related to study treatment; 4 SAEs (2 cases of febrile neutropenia, urinary tract infection,
and anemia) were considered possibly related; 1 SAE (lung infection) was considered remotely
related; and the other 6 SAEs (infection, sciatica, non-cardiac chest pain, dyspnea, wrist
fracture, and neutropenia) were considered unrelated. Three patients died due to disease
progression.

Eleven patients across all groups and schedules had at least 1 event considered a dose
limiting toxicity (DLT): a treatment-related SAE of Grade 3 febrile neutropenia, a
treatment-related SAE of Grade 4 febrile neutropenia, a treatment-related SAE of Grade 4
platelet count decreased, 3 cases of treatment-related SAEs of Grade 4 thrombocytopenia, a
treatment-related AE of Grade 4 thrombocytopenia, 3 cases of treatment-related AEs of Grade 4
thrombocytopenia, a treatment-related AE of Grade 4 neutropenia, a treatment-related SAE of
Grade 4 neutropenia, a treatment-related AE of Grade 3 diarrhea, 2 cases of treatment-related
AEs of Grade 3 nausea, and a treatment related AE of Grade 3 vomiting. Six patients
experienced 8 AEs that led to withdrawal from treatment: 1 (Grade 3 AE of cataracts) was
considered remotely related to study treatment, 4 (Grade 4 AE of thrombocytopenia/Grade 3 SAE
of febrile neutropenia, Grade 2 AE of neutropenia, Grade 3 SAE of febrile neutropenia) were
considered possibly related to study treatment, and 3 (Grade 4 SAE of neutropenia, Grade 4
SAE of thrombocytopenia/Grade 4 SAE of neutrophil count decreased) were considered probably
related to study treatment.

RG7388 is representative of an entirely new branch of the nutlin family of MDM2 antagonists
and the second agent with this mechanism of action sponsored by Roche. As with the lead
compound (RO5045337), RG7388 binds selectively to the p53 site on the surface of the MDM2
molecule in vitro with high affinity and can effectively displace p53 from MDM2, leading to
stabilization and accumulation of p53 protein and activation of the p53 pathway. RG7388 is
from a different chemical series compared to the lead compound and binds with higher potency
and selectivity to the MDM2 protein. This follow-on compound has substantially improved
pharmacological properties. Preclinical models predict superior efficacy of RG7388 in the
clinic at lower doses and exposures. Compared to the lead molecule, lower variability may be
seen given that RG7388 does not have significant pH dependent solubility (fasted and
fed-state simulated intestinal fluid solubilities are similar). RG7388 at 0.3 μM has
equivalent apoptotic activity to RO5045337 at 10-fold higher (3 μM) concentration and at 25
mg/kg has equivalent efficacy to RO5045337 at 4-fold higher dose (100 mg/kg) and better
potency when given once weekly.

RG7388 exhibits improved in vitro and in vivo potency against tumor cell lines and
xenografts, improved CYP inhibition profile, and 2.5- to 20-fold lower projected human
efficacious dose. Preclinical studies have indicated that tumors expressing wild-type p53 may
respond to this novel therapeutic strategy that releases p53 from MDM2 inhibition, and tumors
with wild-type p53 and MDM2 over-expression or amplification are likely to be the most
sensitive. Results from preclinical safety and toxicology studies support further exploration
of this compound in cancer patients. In view of the existing unmet medical need in advanced
cancers expressing the above molecular signature, RG7388 is believed to be a promising agent
that may offer a new therapeutic option.

The use of Pegasys in combination with other targeted agents for the treatment of PV/ET would
allow for the administration of lower doses of Pegasys, thereby reducing dose dependent
toxicities. Preclinical studies by Lu et al. have demonstrated preferential inhibition of PV
CD34+ cell proliferation and colony formation at subtherapeutic doses of Pegasys in
combination with Nutlin-3 [33]. Pegasys increases p53 expression through enhanced
transcriptional activity via the p38 Map kinase pathway and Nutlin-3 prevents p53 degradation
via interruption of the MDM2-p53 interaction [34]. Due to the common downstream target effect
of increased tumor suppressor p53 expression with combination Pegasys and Nutlin-3, lower
doses of each agent can be used together in PV patients. Thus, the use of RG7388 alone or in
combination with Pegasys would be anticipated to enhance p53 expression by different
mechanisms and presents a novel approach to the treatment of PV/ET.

Inclusion Criteria:

(Patient should meet all the criteria)

- JAK2V617F-positive PV or JAK2V617F-positive ET (confirmed by WHO diagnostic criteria)

- High risk ET/PV [age >60; history of thrombosis] or low risk disease with symptoms
[recurrent headaches, paresthesias, pruritus]

- Previously treated with at least one other agent [hydroxyurea, interferon, anagrelide]
and determined to be either intolerant/resistant

-≥18 years of age

- Eastern Cooperative Oncology Group (ECOG) Performance status 0-2

- Acceptable pre-study organ function during screening as defined as: Total bilirubin ≤
1.5 times the upper limit of normal (ULN) unless due to Gilbert's disease or
hemolysis, Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) ≤ 2.5
times ULN, Serum creatinine ≤ 1.5 x ULN

- Women of childbearing potential and males must agree to use adequate contraception
(i.e., hormonal or barrier method of birth control; abstinence) prior to study entry
and for the duration of study participation. Should a female subject become pregnant
or suspect she is pregnant while participating in this study, she should inform the
treating physician immediately

- Ability to understand and willingness to sign a written informed consent document.

Exclusion Criteria:

- Meets the criteria for post ET/PV MF as defined by the International Working
Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT)

- Blast phase disease (>20% blasts in the marrow or peripheral blood)

- Acute thrombosis within 3 months of screening

- Uncontrolled intercurrent illness including, but not limited to hepatitis, human
immunodeficiency virus (HIV) - positive subjects receiving combination antiretroviral
therapy, ongoing or active infection, symptomatic congestive heart failure, unstable
angina pectoris, ventricular arrhythmia, or psychiatric illness/social situations that
would limit compliance with study requirements.