Consequently, following phase I and II clinical trials, imatinib was rapidly tested in a pivotal phase III trial – the International Randomised Study of Interferon versus STI-571 (IRIS) trial – which showed superiority of this compound over the previous standard therapy with interferon-alpha (IFN?). Imatinib provides an effective and durable therapy for newly diagnosed chronic-phase CML, inducing a complete haematological remission in 98% of the patients.

After a five-year follow-up, imatinib resulted in a complete cytogenetic response (CcyR) rate of 87% compared with 69% after one year, and progression into accelerated-phase or blast crisis occurred at an estimated rate of 7%. Additionally, a lower risk of progression was identified in patients with a CcyR or a three log reduction of BCR-ABL transcripts, and recent data are indicating a decreasing annual rate of progression after two, three and four years of therapy. Grade three or four toxicities diminished over time, confirming the excellent tolerability profile compared with IFN.14

Resistance to Imatinib

However, advanced stages of the disease (accelerated phase and blast crisis) as well as chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL) are associated with a significantly decreased response rate with a median overall survival in each of these patient populations of less than one year.15–17

Moreover, not only in advanced CML but also in late chronic-phase patients, a variety of mechanisms of resistance have been observed that eventually lead to cytogenetic or even haematological relapse.

Most importantly, more than 45 point mutations have been identified that interfere with a proper binding of imatinib in the binding domain. Although not readily investigated, the frequency of point mutations depends on the stage of the disease and seems to be the most abundant mechanism of resistance.

Generally, point mutations can occur in both the A- or P-loop, or any other location that ultimately results in a conformational change leading to inadequate or no binding of imatinib.

Depending on the localisation of those mutations within the binding domain dose escalation of imatinib from 400 to 800mg may efficiently overcome resistance. However, mutations located in the catalytic P-loop hardly respond to dose intensification.

Aside from point mutations, the amplification of the Philadelphia chromosome has been identified as the second most common mechanism of resistance. Here, multiple copies of the Philadelphia translocation significantly elevate the cytosolic BCR-ABL concentration, which can be overcome only by dose escalations on a very low level. Finally, the accumulation of additional chromosomal alterations known as clonal evolution may confer resistance through BCR-ABL independent mechanisms.

Aside from dose escalation of imatinib, strategies to respond to clinical resistance may include combination therapies with classical antileukaemic substances, allogeneic stem cell transplantation or the development of second-generation tyrosine kinase inhibitors with improved mechanisms of action. In most cases, improving kinase inhibitors may include the identification of additional targets on the leukaemic clone or the elevation of the binding affinity to the present target structure. However, patients under treatment with imatinib may stop therapy not only because of resistance and relapse of leukaemia but also, in a small but relevant number of patients, because of toxicities. Among these, the most common are fluid retention, muscle cramps, skin rashes and abnormal laboratory parameters. Approximately 25–30% of patients discontinue imatinib because of intolerance.

Nilotinib

Among the various alternative structures that are currently at different stages of pre-clinical and clinical evaluation is nilotinib, Tasigna (formerly AMN107). This is similar to imatinib, and is a 2-phenylaminopyrimidine derivative that has been designed following a rational drug design based on the crystal structure of candidate inhibitors that complex with ABL. The pre-clinical characterisation of this compound with regard to the interaction with the ABL kinase showed an inhibitory concentration (IC) 50 of <30nM (imatinib: 649nM) that corresponds with an approximately 30-fold increase of binding affinity compared to imatinib. Even more favourable is the observation of comparable IC50 with regard to cytokine receptor (KIT) (158nM) and platelet-derived growth receptor (PDGFR) (53nM), the two additional targets of both imatinib and nilotionib.

Pre-clinical testing showed activity of nilotinib in 32/33-point mutants. In addition, nilotinib was shown to have efficacy at well tolerated oral doses in mouse studies for CML. Consequently, a phase I study of nilotinib in 97 patients with imatinib-resistant CML in chronic, accelerated or blast crisis and in nine patients with Ph+ ALL was performed with doses ranging between 50 and 1200mg daily.2 In the chronic-phase patient population of this trial (n=17), complete haematological remissions (CHRs) were demonstrated in 92% and complete cytogenetic remissions in 35%. The nilotinib dose identified for the consecutive phase II trial was 400mg twice a day (BID), which interestingly resulted in plasma concentrations of >1.7 million that are known to effectively inhibit 32/33- point mutation in a cell culture setting.

The most recent phase II data are derived from a multicentre, international trial in over 100 centres with nearly 800 patients.19 Patients enrolled into this trial were suffering either from Ph+ CML or ALL, or were diagnosed for hypereosinophilic syndrome or systemic mastocytosis. CML or ALL patients could be enrolled because of either previous imatinib resistance or imatinib intolerance.

In the chronic-phase population, imatinib resistance was defined as not having achieved a CHR after at least three months of therapy, no minimal cytogenetic remission after at least six months or no major cytogenetic remission (MCyR) after at least 12 months of therapy. Alternatively, patients could be enrolled because of secondary resistance, defined as the loss of a previously achieved CHR or MCyR. All resistant patients had to be pre-treated with at least 600mg/day imatinib for three months. In contrast, the definition of imatinib intolerance is, naturally, less stringent. In this study the definition of imatinib intolerance required that imatinib was discontinued and that patients did not have a baseline MCyR.

Baseline characteristics of the chronic phase patients that were evaluated for toxicity (n=318) showed 70.1% imatinib-resistant versus 29.9% imatinib-intolerant patients. The median exposure to nilotinib was 245 days (range 1–502) and the median dose intensity of 796.6mg/day was only insignificantly below the intended dose of 800mg/day, and is indicative for a relatively good tolerability. Importantly, the chronic-phase study population of this trial included patients with a median age of 58 years (range 21–85) and relatively long disease durations of 57.3 months (range 4.9–275). Therefore, the chronic-phase population of this trial already included intensely pre-treated patients, of whom a substantial proportion were, aside from imatinib, pre-treated with hydroxyurea, INF?, busulfan and systemic chemotherapy.

Efficacy was determined in the first consecutively enrolled 280 patients with at least a six-month follow-up. When only patients without a baseline CHR were considered, the rate of CHR was 74.1%. The MCyR rate in all patients at the same time was 51.8% without any significant difference in patients who were enrolled because of resistance (50.5%) or intolerance (54.7%).

In this relatively short observation time, a median time to progression could not be determined. However, the compound is associated with both a short median time to CHR of one month (range 0.9–8.3 months) and a short median time to MCyR of 2.8 months (range 0.9–11.1).

The toxicity profile of nilotinib is excellent. No clinical grade III/IV nonhaematological toxicities were observed at a frequency >3%. The most common toxicities (grade I/II) were skin rashes, headache and pruritus not requiring dose modifications in most of the cases. However, a considerable number of patients demonstrated grade III/IV biochemistry abnormalities including hypophosphataemia (10%), lipase elevation (15%) and hyperglycaemia (11%). The pathomechanism of these abnormalities is obscure. Nevertheless, normalisation of laboratory parameters occurred in individual patients without dose modifications. In none of the patients was any significant, above grade II fluid retention observed.

In contrast, myelosuppression occurred in a considerable number of patients, reflecting the therapeutic potential of nilotinib. When all grades of myelotoxicity were considered, anaemia occurred in 48.6%, neutropaenia in 49.7% and thrombocytopaenia in 57.3% of all cases. However, severe (grade III and IV) grades of myelotoxicity occurred in only 7.9% (anaemia) to 28.4% (neutropaenia and/or thrombocytopaenia) with a relatively short median duration of fewer than 22 days. Most importantly, nilotinib shows a negligible degree of crossintolerance towards imatinib. In 117 patients with either chronicphase or accelerated-phase CML in this study, cross-intolerance was observed only in one patient with gastrointestinal toxicity and one patient with liver toxicity.

In summary, nilotinib represents a promising novel therapeutic agent in the treatment of imatinib-resistant or -intolerant chronic-phase CML patients. In contrast to other second generation kinase inhibitors in this patient population, nilotinib maintains target specificity but widens the spectrum of sensitivity, as a large number of known BCR-ABL mutants are sensitive to this substance.