Abstract 113

Introduction.

Adherence to therapy is known to be a critical factor for achieving an optimal imatinib (IM) response, and non-adherence is likely to be a cause of loss of response. In the absence of a biological marker of resistance, such as BCR-ABL mutations or blast crisis (BC), it is difficult to determine whether loss of response is associated with non-adherence or drug resistance. Prior studies found that the rate of a BCR-ABL rise provided biological insight into the disease phase at relapse (Blood 1996 87 4473; Blood 2004 104 2926). A more rapid rate of rise occurred in patients (pts) who relapsed into advanced phases compared to those who relapsed into chronic phase (CP), as assessed by the number of days over which BCR-ABL doubled (doubling-time, DT). We compared the DTs of pts with progression to BC and those who acquired a BCR-ABL mutation and maintained CP to the DTs of responding pts with documented IM discontinuation/interruption. The aim was to determine whether IM interruption was associated with characteristic BCR-ABL kinetics and loss of response, and whether the DT could help to differentiate drug resistance from non-adherence.

Method.

The molecular data was examined for 179 CP pts who achieved a major cytogenetic response and met these inclusion criteria: relapsed directly into BC at the time of a significant BCR-ABL rise; or a mutation was detected and CP was maintained at the time of a rise; or discontinued IM while in a sustained complete molecular response (CMR) as part of a discontinuation study; and/or IM dose was documented for BCR-ABL measurement intervals. IM dose interruption was defined as zero IM for ≥10% of days of a BCR-ABL measurement interval.

Results.

Twelve pts relapsed directly into BC and their BCR-ABL DT was short (median 9.0 days (d), range (r) 6.1–17.6); 8 had mutations and there was no DT difference for these pts. In contrast, 28 pts who maintained CP after mutation detection had a longer DT: median 45 d (r 17–140), P<.0001 (Figure). These pts met one or more criteria for IM failure (only 1 subsequently progressed to BC on IM after 8 months).

While in CMR, 36 pts discontinued IM and 17 had a significant BCR-ABL rise and molecular relapse. The DT (median 9.0 d, r 6.9–27) was similar to the BC pts, P=.52. The short DT of pts who discontinued IM is consistent with the rate at which terminally differentiated leukemic cells arise from leukemic progenitors in the absence of IM (Nature 2005 435 1267, 8 d DT). The similarly short DT of BC pts suggests complete lack of BCR-ABL inhibition by IM, whereas relapse into CP with mutations was characterized by slow BCR-ABL kinetics, presumably due to maintenance of partial BCR-ABL inhibition by IM.

We determined whether temporary IM dose interruption in responding pts also resulted in a BCR-ABL rise and characteristic DTs. IM interruption occurred in 42 pts (46 interruptions) for: intolerance 32; non-adherence 8; second malignancy/surgery 3; other 3. The IM interruptions were divided into 2 groups according to the days of zero IM: 100% of BCR-ABL measurement interval days (complete lack of BCR-ABL inhibition); or 10 to <100% of days (partial BCR-ABL inhibition). Twelve pts had 100% IM interruption and all had a rapid BCR-ABL rise and short DTs (median 9.4 d, r 4.2–17.6), consistent with the kinetic pattern of BC and discontinued in CMR pts, P=.78. Loss of response occurred in 11/12. Thirty pts had 34 measurement intervals of zero IM for 10 to <100% of days (median 33%, r 10–89). A BCR-ABL rise occurred in 27/34 interruptions and the DT (median 26 d, r 11.6–87) was significantly longer than BC, discontinued in CMR and 100% interruption pts, P<.0001. BCR-ABL remained stable in 5/34 interruptions and a reduction occurred in 2. Loss of response occurred in 18/34 interruptions. Of all 46 measurement intervals with an IM interruption (complete or partial), a significant BCR-ABL rise occurred in 39 (85%) and loss of response in 29 (63%), including 7/8 interruptions due to non-adherence. No pt progressed during the interruption. Twenty-one pts with a mutation and/or BC relapse had dose data available: none had an IM interruption during the relevant measurement interval.

Conclusion.

IM interruption is associated with a BCR-ABL rise. Intermittent IM dosing may be associated with a longer DT, similar to that seen with the emergence of mutations in CP. A short BCR-ABL DT is typical of complete loss of BCR-ABL inhibition and, for a patient still in CP, should raise the suspicion of non-adherence.

Disclosures:

Branford:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Bristol Myers Squibb: Honoraria, Research Funding; Ariad: Research Funding. Yeung:Novartis: Research Funding; Bristol Myer Squibb: Research Funding. Ross:Novartis: Honoraria, Research Funding. Hughes:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Ariad: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution