Chronic Graft versus Host Disease (cGvHD) has been identified as the leading cause of late non-relapse mortality in Hemopoietic Stem Cell Transplant (HSCT) survivors. Up to now a standard satisfactory treatment for these patients does not exist. cGVHD is an immune-mediated disease, resulting from a complex interaction between donor and recipient adaptive immunity, but its exact pathogenesis is still incompletely defined.

The purpose of this study is to determine safety and efficacy of Nilotinib in a population with steroid-refractory/or steroid-dependent cGvHD with a phase I study.

In phase II the MTD will be used to define the efficacy of Nilotinib in a cGvHD steroid- refractory or steroid dependent population, with the same characteristics of the previously Imatinib-treated population.

Chronic Graft versus Host Disease (cGvHD) has been identified as the leading cause of late non-relapse mortality in Hemopoietic Stem Cell Transplant (HSCT) survivors. Up to now a standard satisfactory treatment for these patients does not exist. cGVHD is an immune-mediated disease, resulting from a complex interaction between donor and recipient adaptive immunity, but its exact pathogenesis is still incompletely defined.

Many cGVHD patients show skin and visceral involvement with fibrotic scleroderma-like features; recent experimental data suggest that patients with cGVHD with fibrotic/sclerotic manifestations, have agonistic antibodies activating the platelet-derived growth factor receptor (PDGF-R), like patients with Systemic Scleroderma (SS), suggesting a similar pathogenesis for these two diseases; recently it has been demonstrated that the up-regulation of the PDGF-R intracellular pathway leads to increased reactive oxygen species (ROS), with consequent increased collagen synthesis, which, in turn, contributes to the pathologic lesions observed in both cGVHD and SS.

Many data strongly suggest that, among the profibrotic cytokines, besides PDGF, also transforming growth factor β (TGF-β) can play a relevant role in the pathogenesis of fibrotic damage. Blockade of PDGF or TGF-β signaling has been shown to reduce the development of fibrosis in various experimental models. Recent in vitro data showed that the Tyrosine Kinase Inhibitor (TKI), Imatinib strongly inhibits both PDGF and TGF-β intracellular signaling. Considering this selective, dual inhibition by Imatinib of the TGF-β/PDGF pathways, we conducted a pilot study evaluating safety and activity of Imatinib in a series of patients affected by refractory cGVHD, with fibrotic/sclerotic features.

In 2008 we started a national pilot multicentric study (supported by AIFA) to confirm these preliminary data; until now, 24 patients have been enrolled, and an interim analysis confirmed that Imatinib at low dosage is safe and well tolerated in this setting; however the response rate observed in this second study is 50% at sixth month, apparently inferior to the first pilot experience. Two possible reasons could be responsible for this discrepancy: first, in the last study, the response criteria are more stringent, being related to a multi-organ evaluation (according to the NIH Consensus Conference on GVHD); secondly, the average age of the patients enrolled in the second trial is significantly superior to the other ones (in the first study we observed the best responses in young patients). It's important to point out that the higher tolerated dose of Imatinib in this setting of patients is significantly inferior (100-200 mg) to the dosage currently used in patients with Chronic Myeloid Leukemia.

Like Imatinib, Nilotinib is a TKI with better tolerability profile suggesting a better compliance, especially in frail patients with poor haematological tolerance and poor Quality of Life (QoL), such as those affected by steroid-refractory cGvHD.

Therefore in the population of allotransplanted patients, with steroid-refractory cGvHD, an innovative approach with a novel TKI is justified for the following reasons:

1. Both first-generation TKI and second-generation TKI have been largely used for long periods in many patients with malignancies, showing a good safety profile;
2. Imatinib achieved preliminary encouraging results in refractory cGvHD patients, who represent a frail and truly orphan population with a very poor quality of life;
3. The availability of a new TKI with better tolerability profile can achieve not only a better response rate, (which may be considered perhaps a cosmetic result), but it could be also better tolerated for long time, allowing to improve the compliance and the duration of treatment in responding patients, with a consequent impact on the long term outcome of the transplantation procedure.

Nilotinib is a novel aminopyrimidine, protein-tyrosine kinase inhibitor which has been developed based on crystallographic studies of the interaction between IM (imatinib) and ABL (tyrosine kinase product of Abelson gene), so as to increase the affinity and the binding capacity of the derivative to wild type ABL and more important, to the ABL mutants which are the main cause of IM resistance.

Nilotinib is active against the same TK as Imatinib, namely Abl, BCR-ABL (BCR-ABL oncoprotein product of BCR-ABL fusion gene), PDGFR-α and β, and c-Kit, but it is at least 10 times more potent against BCR-ABL.

Moreover, Nilotinib is not active, like Imatinib, against several other TKs, including VEGFR-2, c-ERB-2, FLT3 (fetal liver tyrosine kinase 3), IGFR-1R (insulin-like growth factor receptor 1), Jak-2, RAS, lk and AKT. More interesting and more important (Table 2), Nilotinib is more active that Imatinib against a number of BCR-ABL mutants, with the exception of the T315I. It has to be noticed that in all the proliferation assays the concentration at which 50% of effect is inhibited of Nilotinib is lower than 50 nM, apart from the two BCR-ABL mutants (E255K and E255V) where the IC50 is 600 to 700 nM.

We have design a prospective, phase I/II, non-randomized, open-label, multicenter study to determine safety and efficacy of Nilotinib in a population with steroid-refractory/or steroid-dependent cGvHD. In phase I approximately 3-24 patients will be needed to find the MTD depending on the toxicities seen.

In phase II the MTD will be used to define the efficacy of Nilotinib in a cGvHD steroid- refractory or steroid dependent population, with the same characteristics of the previously Imatinib-treated population.

STATISTICAL DESIGN AND SAMPLE SIZE Phase I (Dose finding). Phase I will be conducted according to a standard 3+3 design. Initially, three patients will receive the lowest dose of Nilotinib, 200 mg QD. If no dose-limiting toxicity (DLT), defined as that dose which includes any Grade III/IV toxicity by the NCI-COMMON TOXICITY CRITERIA criteria is observed in the current cohort after at least one month of treatment, the next three patients will be treated with the next higher dose of Nilotinib (300 mg QD), then 200 mg BID, then 300 mg BID).

If two or more DLTs are observed, the Phase I will end and MTD will be defined as the next lower dosage. If we will observe one DLT (1/3 patients), we will treat other three patients with the same dosage; if there isn't another DLT in these three patients, we will increase the dosage in the next cohort; but if we will observe another DLT in this cohort (total DLT 2/6 patients), this means that we exceeded the MTD, and the study will continue in phase II using the previous dose. MTD will be considered the maximum dosage used to treat six patients achieving at most one DLT, or the minimum dosage if with this dosage we will achieve two DLT. In case of limiting Grade II toxicity a meeting of the data safety monitoring board (DSMB) might be held to decide how to continue the study. Pharmacokinetics: pharmacokinetics serum levels of Nilotinib will be determined in all patients within this phase by using mass-spectrometry assay, thanks to collaboration of Professor F. Pane, Naples.

Doses will not be escalated beyond 600 or below 200 mg/die. Then the algorithm will proceed into Phase II. The dose estimated as the MTD in phase I will be used for phase II, this dose will be reviewed and may be adjusted accordingly by the DSMB after the analysis of toxicity obtained from all phase I patients at least 3 months under therapy.

Phase II (Clinical efficacy part) To evaluate the efficacy of Nilotinib treatment the study will follow a Minimax design. This design consists in two subsequent enrolment phases. 19 patients will be recruited in the first stage and, if after the enrolment not more than 6 responses will happen, then the trial will end. In the second stage, other 20 patients will be recruited. The upper limit of the second stage rejection of the drug is 16.

The whole number of recruited patients will be 39 with an early termination probability of 67%. If after the first stage, the number of responses is fewer than or equal to a pre-determined value, r1 of n1, then PET is equal B(r1;n1,p) where B denotes the cumulative distribution function evaluated at r1 for a random variable that is binomially distributed with parameters n1 and p is the true probability of response. If the trial continues to the second stage, then the drug is rejected at the end of this stage if r or fewer responses are observed.

Based on the best results observed in the literature with experimental treatments of steroid refractory cGvHD, we choose an overall minimal response (CR+PR) rate at least 50%, that is p1=0.50. The rate of ineffectiveness after the first stage is required not greater than 30%, (no attention is paid to its correlation to the disease or to the treatment), so p0=0.30. If the significance level is alpha=0.05, requiring a power of 1-beta=0.80, they have n1=19, r1=6 according to the tables.

The study will follow the Good Clinical Practice rules, and the ongoing trial will be monitored by the promoter GITMO "Trials Office".

Primary objective in Phase I The primary objective of Phase I is to evaluate the safety of Nilotinib and defining the MTD of the drug.

Primary objective in Phase II:

In phase II the MTD will be used to define the efficacy of Nilotinib in a cGVHD steroid refractory population. In this phase, the primary end-point is the Overall Response Rate (ORR), defined as objective Improvement at sixth month, according to modified NIH Consensus Conference criteria.

Secondary objectives:

Reduction or stopping the concomitant immune-suppressive treatment (steroids, Extra-corporeal Photopheresis, Cyclosporin-A).

Evaluation of Nilotinib tolerance along the 6 months of therapy: incidence of Serious Adverse Event (grade>2 according to WHO scale) in patients with refractory cGVHD -Evaluation of Nilotinib activity in patients with: skin fibrosis; lung fibrosis: Bronchiolitis Obliterans Organizing Pneumonia (BOOP) and/or pulmonary hypertension; visceral fibrosis. Appropriate assessment attachments will be used to measure the response in these districts, according to the cGVHD assessment worksheet of NIH Consensus Conference Documents and the modified Rodnan Score.

TTF (Time to Treatment failure). OS at 1 year after start of Nilotinib therapy. Global Health state evaluated by Health Assessment Questionnaire The cGVHD symptom scale Biological objectives

To evaluate the relationship between the serum levels of Nilotinib and response:

Evaluation of the immunomodulatory effect of Nilotinib on T populations. To evaluate the relationship between response to Nilotinib and anti-PDGF-R antibody activity/titer Analysis of expression of selected auto-antibody complementarity-determining regions (CDR) in RNA of cGVHD patients.

Evaluation of role of PDGF-R and TGFB-R pathway activation in cGVHD patient's with skin sclerosis.

To evaluate the fibroblast activation pathway modifications before and after Nilotinib administration in vitro: decrease or de-phosphorylation of fibroblast activation markers (ROS, Ha-Ras, extracellular signal-regulated kinase 1/2, collagen type I, Smad, C-Abl) as index of PRGF-R/TGFB-R pathway inactivation (evaluated both in patient's fibroblast from skin biopsy and in immortalized fibroblast) after Nilotinib treatment.

To set up a reliable and easy technique (alternative to the functional test used in the previous experimental works, based on the ROS output from murine fibroblasts, incubated with the patients purified immunoglobulins) for screening patients for the presence of stimulating anti-PDGFR antibodies before and after Nilotinib treatment.

Inclusion Criteria:

• Written informed consent
• Male
• not pregnant female
• patients >=18 and <65 years old
• Weight >40 Kg
• Fertile female must use both anti-conception devices and oral contraceptives
• Diagnosis of cGVHD steroid refractory (no response after Prednisone ≥1mg/kg along 6-8 weeks) or steroid-dependent cGVHD (need of > 0.4 mg/Kg/die of Prednisone continuously)
• Patient intolerant to steroid therapy
• Patients with extensive cGVHD including one of the following features:
• skin sclerosis in more than 50% body surface area; active disease with significant progression in the last 6 months or
• skin sclerosis in less than 50% BSA, but presence of visceral involvement or
• Lung cGVHD involvement, documented by Histology (when possible) and/or High Resolution computed tomography scan plus significant alterations of Respiratory tests: forced vital capacity or diffusion capacity deterioration in the last 12 months; Forced expiratory volume in one second <75% predicted ratio within 1 year; evidence of air-trapping or small-airway thickening or bronchiectasis on High-resolution computed tomography or pathologic confirmation of constrictive bronchiolitis; no evidence of active infection in the respiratory tract, documented with investigations directed by clinical symptoms, including radiologic studies or microbiologic cultures. A quantitative lung involvement by cGVHD should be made by using the modified Lung Functional Score* (LFS).
• Visceral sclerosis clinically relevant with digestive involvement also without skin involvement; biopsy at physician discretion.
• In all patient with skin involvement the cGVHD should by documented by Histology
• Patients with visceral involvement clinically and technically documented, but without skin sclerosis will be included if the clinical diagnosis of cGVHD is conformed to NIH criteria
• LFS calculated according to NIH consensus project on criteria for clinical trials in cGVHD
• Failure of at least two immunosuppressive lines, including the steroids
• Lab criteria:
• Alanine aminotransferase and aspartate aminotransferase <2.5 x Upper Limit of Normal or >5.0 x Upper Limit of Normal if considered due to the disease Alkaline phosphatase <2.5 x Upper Limit of Normal
• Serum bilirubin <1.5 x Upper Limit of Normal
• Serum creatinine <1.5 x Upper Limit of Normal
• Serum amylase <1.5 x Upper Limit of Normal and serum lipase <1.5 x Upper Limit of Normal
• Normal serum level of potassium, total calcium corrected for serum albumin; magnesium and phosphorus
• Absolute neutrophil count≥1000/mmc
• Platelets ≥50,000 mmc

Exclusion Criteria:

• Patients with stable disease, well controlled by the current treatment
• Patients who do not need high-dose steroids (daily dose of prednisone <0.4 mg/kg/day) and/or other immunosuppressive agents
• Pregnancy, fertile female without intention to use contraceptives or breast feeding
• Previous treatment with Imatinib or Rituximab in the last six months
• Severe liver or renal impairment: serum creatinine >2,5 mg/dl; serum bilirubin>2,5 mg/dl (without evidence of hepatic cGVHD)
• Other uncontrolled malignancies including the persistence of the underlying malignancy before the Allogeneic Transplantation.
• Any other investigational agents administered within last four weeks
• History of myocardial infarction within the last 12 months
• Uncontrolled angina pectoris
• Cardiac insufficiency (>grade II, New York Heart Association classification)
• Arrhythmia
• Long QT syndrome and/or corrected QT interval >450 msec on screening ECG
• History of acute or chronic pancreatitis
• Use of therapeutic coumarin derivates
• Other concurrent severe and/or uncontrolled medical conditions that could cause unacceptable safety risks or compromise compliance with the protocol
• Use of all strong CYP3A4 inhibitors is excluded.