ABL Mutations in Late Chronic Phase Chronic Myeloid Leukemia Patients With Up-Front Cytogenetic Resistance to Imatinib Are Associated With a
http://www.100md.com
《临床肿瘤学》
the Institute of Hematology and Medical Oncology "Seràgnoli," University of Bologna, Bologna
CEINGE Advanced Biotechnologies and Department of Biochemistry and Medical Biotechnology, University of Naples "Federico II," Naples
Division of Hematology and Internal Medicine, Department of Clinical and Biological Sciences, University of Turin, Turin
Novartis Pharma, Origgio, Italy
ABSTRACT
PURPOSE: Point mutations within the ABL kinase domain of the BCR-ABL gene have been associated with clinical resistance to imatinib mesylate in chronic myeloid leukemia (CML) patients. To shed further light on the frequency, distribution, and prognostic significance of ABL mutations, we retrospectively analyzed a homogeneous cohort of late chronic phase CML patients who showed primary cytogenetic resistance to imatinib.
PATIENTS AND METHODS: Using denaturing high-performance liquid chromatography (D-HPLC) and sequencing, we screened for ABL mutations in a total of 178 bone marrow and/or peripheral blood samples from 40 late chronic phase CML patients homogeneously treated with imatinib 400 mg/d, who did not reach a major cytogenetic response at 12 months.
RESULTS: Mutations were found in 19 of 40 patients (48%). Mutations were already detectable by D-HPLC at a median of 3 months from the onset of therapy. The presence of a missense mutation was significantly associated with a greater likelihood of subsequent progression to accelerated phase/blast crisis (P = .0002) and shorter survival (P = .001). Patients carrying mutations falling within the P-loop seemed to have a particularly poor outcome in terms of time to progression (P = .032) and survival (P = .045).
CONCLUSION: Our results show that, irrespective of the hematologic response, monitoring for emerging mutations in the first months of therapy may play a role in detecting patients with worse prognosis, for whom a revision of the therapeutic strategy should be considered.
INTRODUCTION
The exciting results from trials of imatinib mesylate in patients with chronic myeloid leukemia (CML) have established it as the new standard of care for the chronic phase (CP) and accelerated phase (AP) of the disease.1-4 Imatinib is a potent and selective inhibitor of the BCR-ABL tyrosine kinase, which is known to be deregulated in as many as 95% of CML patients. However, despite high rates of hematologic and cytogenetic responses, primary refractoriness and acquired resistance have been observed in both late CP and AP patients. Point mutations within the ABL kinase domain have been reported as the most frequent mechanism for BCR-ABL reactivation within the leukemic clone.5-12 Structural data suggest that imatinib acts by blocking the BCR-ABL protein into the inactive conformation.13,14 This prevents the transfer of phosphate from adenosine triphosphate (ATP) to substrates and blocks the downstream signal transduction pathways.15-17 Mutations seem to act by disrupting critical contact points between imatinib and BCR-ABL or, more often, by inducing a transition from the inactive to the active state, a conformation to which imatinib is unable to bind.18,19
analyses of clinical samples, the repertoire of mutations found in association with the resistant phenotype has been accruing slowly but inexorably over time, with an overall frequency ranging from 26% to 90% of patients.5-12 Such a difference primarily reflects heterogeneity in patient populations. Most of the studies examined mixed populations of both Philadelphia-positive acute lymphoblastic leukemia and CML patients, the latter at different stages of disease. The criteria for patient selection and the definitions of resistance were also different from one study to another. Most of the studies have focused on patients experiencing disease relapse while receiving imatinib therapy, whereas few data are available to date on the incidence of mutations in patients with primary resistance to imatinib.10,12
Another issue that deserves additional elucidation is the prognostic significance of ABL mutation detection. Shah et al10 screened for mutations in 13 cytogenetic nonresponders and found that three of four patients with mutations experienced disease progression within 18 months from mutation detection. In contrast, only one of nine patients without a detectable mutation experienced disease progression. Hochhaus et al12 examined 43 patients with hematologic relapse and did not find any difference between patients with and without mutations regarding time to progression. Branford et al20 recently showed that in late CP and AP CML patients, a specific subgroup of mutations (ie, those in the ATP phosphate-binding loop [P-loop]) are significantly associated with a poor prognosis in terms of survival.
We recently described a novel denaturing high-performance liquid chromatography (D-HPLC) -based assay for screening of ABL kinase mutations.21 To shed further light on the frequency and prognostic significance of ABL mutations in CML patients resistant to imatinib, we have applied this method to retrospectively analyze a homogeneous cohort of 40 late CP CML patients resistant or refractory to interferon alfa (IFN-) enrolled onto the CML/002/STI571 multicenter clinical trial of the Gruppo Italiano Malattie Ematologiche dell’Adulto (GIMEMA) Working Party on CML (formerly the Italian Cooperative Study Group on CML),22 who showed up-front cytogenetic resistance to imatinib.
Our results confirm that ABL mutations are a relevant mechanism of resistance also in the subset of late-CP patients who are cytogenetically refractory to imatinib. In addition, we found that mutations are significantly associated with a greater likelihood of progression to AP/blast crisis (BC) and death as a result of disease, and that mutations falling within the P-loop confer a particularly poor prognosis.
PATIENTS AND METHODS
Patients and Experimental Design
The study was performed on a total of 178 bone marrow (BM) and/or peripheral blood (PB) samples obtained from 40 patients enrolled onto the CML/002/STI571 multicenter clinical trial of the GIMEMA Working Party on CML.22 The trial enrolled 191 CML patients between June and December 2000. To be eligible, patients were required to be in first CP and to be either intolerant of or resistant to previous IFN- treatment. Patients resistant to IFN- were divided into two groups: hematologic-resistant patients (ie, those who failed to achieve or to maintain a complete hematologic response with IFN-) and cytogenetic-resistant patients (ie, those who failed to achieve or to maintain a complete or partial cytogenetic response with IFN-).2,22 All of the patients provided signed informed consent for participation in the study. Treatment consisted of imatinib 400 mg once daily until disease progression. If hematologic toxicity grade 3 or 4 occurred, the treatment was discontinued until the toxicity had resolved to grade 2 or less. If nonhematologic toxicity grade 2 or 3 occurred, the treatment was discontinued until the toxicity had resolved to less than grade 2, and was resumed at 400 mg if toxicity was grade 2 and at 300 mg if toxicity was grade 3. If grade 4 nonhematologic toxicity occurred, imatinib was permanently withheld.22
The patients who entered onto this molecular study were part of a larger series (n = 56) who showed primary cytogenetic resistance to imatinib therapy (ie, they failed to obtain at least a major cytogenetic response [MCgR] at 12 months). The selection of 40 patients for this study was exclusively based on the availability of samples at our institution. The median time from diagnosis to the start of imatinib therapy was 3.7 years (range, 1.3 to 16.3 years). Sixteen of 40 patients (40%) permanently discontinued imatinib because of disease progression (10 patients) or toxicity (six patients). After imatinib discontinuation, one patient underwent an allogeneic transplantation. The remaining 15 patients were managed with conventional chemotherapy (mainly hydroxyurea) until the time of last contact or death.
ABL mutational screening was done either on the sample(s) collected at month 12 of therapy for those patients who regularly completed the first year of treatment (n = 32), or on the sample(s) collected immediately before imatinib discontinuation for those patients who exited the protocol before month 12 because of disease progression (n = 5) or toxicity (n = 3). Exceptions were patients 35, 36, 37, and 38, who were analyzed at month 6, 6, 9, and 9 of therapy, respectively, because the samples were unavailable at the subsequent time points. Whenever a mutation was detected, its presence was retrospectively traced on the samples collected at the previous time points, when available, back to the pretreatment sample.
Definitions
The cytogenetic response was based on the evaluation of a minimum of 20 marrow cells. MCgR was defined as more than 65% Philadelphia-negative metaphases. Hematologic response was defined as complete (CHR) if the leukocyte count was less than 10 x 109/L, if no immature cells were recorded in the differential count, if the platelet count was less than 450 x 109/L, and if the spleen was not palpable. AP was defined as a percentage of blasts in PB or BM 15% but less than 30%, a percentage of blasts plus promyelocytes in PB or BM 30%, a percentage of peripheral basophils 20%, or thrombocytopenia less than 100 x 109/L unrelated to therapy. BC was defined as 30% blasts in PB or BM.
RNA Extraction and Reverse Transcriptase Polymerase Chain Reaction
Total cellular RNA was extracted from mononuclear cells obtained by Ficoll-Hypaque density gradient centrifugation and resuspended in guanidinium thiocyanate as previously described.23 RNA was spectrophotometrically quantified at 260 nm and its integrity was assessed by electrophoresis on 2% agarose gel. One microgram of total cellular RNA was reverse transcribed to cDNA using 5 μmol/L of random hexamer primers and 200 U of M-MLV reverse transcriptase (GeneAmp RNA PCR Kit, Applied Biosystems, Foster City, CA).
D-HPLC Analysis
For ABL kinase domain amplification, we used a nested polymerase chain reaction (PCR) approach as previously reported.21 All PCR products were analyzed using a D-HPLC Wave 3500HT DNA Fragment Analysis System (Transgenomic Ltd, Crewe Cheshire, United Kingdom). Sample preparation and elution conditions have been described previously.21 One wild-type sample was used as a negative control. Chromatograms from each tested patient were overlaid with the wild-type profile, and samples with extra peak(s) were scored as positive. To ensure that homozygous mutations could not escape D-HPLC detection, for all samples to be studied a mixture of a wild-type PCR product and patient PCR product in a 1:1 ratio was also run. Sensitivity and reliability of the D-HPLC assay were accurately determined. To assess the sensitivity of the D-HPLC assay, which has been reported to be variable in a range of 1% to 10% depending on the sequence and length of the fragments to be analyzed, limiting dilution experiments were performed by mixing known quantities of amplified wild-type and mutant (G250E, Y253F, E255K, T315I, M351T, H396R) ABL PCR products. The percentages of mutant with respect to wild-type were 50:50% –40:60% –30:70% –25:75% –20:80% –17:83% –15:85% –12:88% –10:90% –7:93% –5:95% –2:98% –1:99%. Results indicated that D-HPLC can reach a lower detection limit between 2% and 5% for the following mutations: G250E, Y253F, E255K, T315I, and H396R, and between 5% and 7% for the M351T ABL mutation.21 To rule out the possibility of false-positive and/or false-negative results, 51 BM and/or PB samples from 27 CML patients were analyzed in parallel by D-HPLC and sequencing. Whenever D-HPLC analysis showed an abnormal elution profile, sequence analysis confirmed the presence of a point mutation. Conversely, all of the samples scored as wild-type by D-HPLC did not show evidence of mutations by direct sequencing.21
Direct Sequencing
Direct sequencing was done using the Big Dye Terminator Cycle Sequencing Kit (Applied Biosystems) and an ABI PRISM 377 DNA Analyzer (Applied Biosystems), as previously reported.21 Sequences were compared with the wild-type sequence using BLAST (ABL accession number X16416). Sequence analysis was performed on both strands for each fragment.
Cloning
In three patients for whom direct sequencing failed to detect any point mutation, PCR products were cloned into a pCR2.1-TA vector (TOPO TA Cloning Kit; Invitrogen, Carlsbad, CA) according to the manufacturer's instructions, and 20 independent clones were sequenced.
Statistical Analysis
To test for differences between patients with mutations and patients without mutations, the Mann-Whitney U test was used for continuous variables and the Fisher's exact test or 2 test was used for categoric variables, as appropriate. Curves for survival and time to progression were estimated using the Kaplan-Meier method. The log-rank test was used to test for differences in survival between groups. All analyses were performed using the Statistical Package for the Social Sciences (SPSS software package; SPSS Inc, Chicago, IL).
RESULTS
D-HPLC and Sequence Analyses for ABL Kinase Mutations
In 19 of 40 patients (48%), D-HPLC analysis showed an abnormal elution profile suggesting the presence of one or more sequence variations. Direct sequencing confirmed the presence of a point mutation in all cases. Results of D-HPLC and sequence analyses are listed in Table 1. A total of 13 different point mutations were detected, all but one leading to an amino acid substitution. Nine patients showed mutations (G250E, Y253F, Y253H, E255K, E255V) falling within the nucleotide-binding loop (P-loop). Ten patients showed mutations outside the P-loop, in other regions of the kinase domain (M244V, F311L, F317L, M351T, E355G, F359V, H396R, and a silent mutation at codon 298).
The 19 patients who had a mutation were then subjected to a longitudinal analysis to trace the presence of the mutation in the available samples collected and stored at previous time points, back to the pretreatment sample. Results are reported in Figure 1. Mutations were already detectable at a median of 3 months (range, 1 to 6 months) from the onset of imatinib therapy. In three patients, D-HPLC analysis of the samples collected after 1 (patient 17) and 2 months (patients 5 and 10) from imatinib onset yielded an abnormal elution profile but the presence of the mutation (F359V, Y253F, and E255K, respectively) was revealed only after cloning in two of 20, three of 20, and three of 20 clones, respectively. D-HPLC analysis did not show evidence of mutations before imatinib initiation in any of the patients.
Correlations Between Point Mutations, Baseline Features, and Outcome
Baseline features of patients with mutations versus patients without mutations are listed in Table 2. Mutational status and clinical outcome for each of the 40 patients analyzed are listed in Table 3. As shown in Table 2, no differences emerged between patients with a missense mutation and patients without a missense mutation regarding sex, median age at the time imatinib treatment was started, and disease history (intolerance/hematologic resistance/cytogenetic resistance to IFN-). Median time from diagnosis to the start of imatinib therapy was 4.8 years for patients with missense mutations and 2.9 years for patients without missense mutations; the difference was borderline statistically significant (P = .056).
The CHR rate was similar in the two groups of patients: at 3 and 6 months, the CHR rate was 83% (15 of 18) and 89% (16 of 18), respectively, for patients with missense mutations, compared with 82% (18 of 22) and 90% (20 of 22), respectively, for patients without missense mutations. At the time of mutation detection, all but four patients with missense mutations were in CHR. The median follow-up was 32.5 months (range, 10 to 36 months). As shown in Table 3, 11 of 18 patients with ABL missense mutations experienced disease progression to AP or BC after a median of 9 months (range, 6 to 30 months) from the time mutation was first detected and a median of 12 months from the onset of imatinib therapy (range, 7 to 33 months); seven of them subsequently died as a result of their disease. In contrast, only two patients of 22 without ABL missense mutations experienced disease progression to AP or BC and none of them has since died (Table 3). Curves for time to progression and survival of patients with a mutation versus patients without a mutation are shown in Figures 2 and 3, respectively. Patients with ABL mutations had a significantly shorter time to progression (log-rank P = .0002) and survival (log-rank P = .001).
Among patients with missense mutations, eight of nine patients with mutations falling within the P-loop (G250E, Y253F, Y253H, E255K, E255V) experienced disease progression to AP or BC after a median of 9 months from the time the mutation was first detected (range, 6 to 29 months) and 12 months from the onset of imatinib (range, 10 to 32 months); six of them subsequently died as a result of their disease. In contrast, only three of nine patients with mutations outside the P-loop experienced disease progression to AP or BC, and only one patient died. Curves for time to progression and survival of patients with and without P-loop mutations are shown in Figures 4 and 5, respectively. Patients with P-loop mutations had a significantly shorter time to progression to AP/BC (log-rank P = .032) and survival (log-rank P = .045).
DISCUSSION
The aim of the present study was to examine the frequency, distribution, and prognostic relevance of ABL kinase domain mutations in a homogeneous cohort of late CP CML patients enrolled onto the CML/002/STI571 multicenter clinical trial.22 Taking advantage of a recently validated D-HPLC-based assay for ABL mutational screening,21 we analyzed 40 patients who showed up-front cytogenetic resistance to imatinib, defined as a failure to reach an MCgR at 12 months. Our results show that ABL kinase mutations are a relevant mechanism of resistance to imatinib also in the poorly investigated setting of primary cytogenetic resistance: we detected a total of 13 different point mutations, all but one leading to an amino acid substitution, in 48% of our patients. Nine patients showed missense mutations falling within the nucleotide-binding loop (P-loop; G250E, Y253F, Y253H, E255K, E255V). Ten patients showed missense mutations in other regions of the kinase domain (M244V, F311L, F317L, M351T, E355G, F359V, H396R). All of these mutations have been described previously.5-12,20,24 In a patient from our series we also found a single amino acid substitution leading to a silent mutation at codon 298. The mutated clone coexisted with the wild-type one at an approximate relative ratio of 50% to 50%. The patient was investigated further to trace the presence of the mutation in the available samples collected and stored at previous time points, back to the pretreatment sample (Fig 1). The nucleotide change was detectable at 6 months and in all the samples collected thereafter, but not in those collected before the onset of therapy and at 1 and 2 months. These observations raise a question regarding the biologic significance of this amino acid substitution. Because mechanisms of resistance other than mutations exist, it may be speculated that the mutated BCR-ABL allele, not conferring per se any selective advantage, originated in a subclone bearing other abnormalities actually responsible for the selection and/or expansion of the subclone itself and for the resistant phenotype.
The same longitudinal, retrospective analysis was performed on all of the other patients bearing mutations (Fig 1) to determine how early during treatment D-HPLC could reveal the mutation, thus evaluating whether our screening method could be predictive of an emerging resistance. Overall, mutations were already detectable by D-HPLC at a median of 3 months (range, 1 to 6 months) from the onset of therapy. All patients but four had sustained hematologic response at the time of mutation detection. Bearing in mind that all the patients herein reported did not reach an MCgR at 12 months, our results indicate that as many as 50% of patients who failed to achieve cytogenetic control of the disease in the first 6 months of therapy with imatinib already had evidence of a point mutation by D-HPLC analysis. These data would suggest that, irrespective of the hematologic response, regular monitoring for emerging mutations in the first months of therapy might help to identify patients for whom a revision of the therapeutic strategy should be taken into account.
Despite allowing the identification of resistance-associated mutations as early as 1 month after the onset of imatinib, D-HPLC screening did not detect mutations in any of the pretreatment samples analyzed. It has been reported recently by some authors that, in some CML and Philadelphia-positive acute lymphoblastic leukemia patients, specific mutations (E255K, F311L, T315I) could be found in a small proportion of leukemic cells before the onset of imatinib treatment.9,24,25 The reason why other studies12,20 failed to detect mutations in pretreatment samples might be a lack of sufficient sensitivity of the techniques used for mutational screening. D-HPLC is a reverse-phase ion pair HPLC specifically developed for detection of DNA sequence variations such as point mutations, small insertions, and deletions.26 Under conditions of partial heat denaturation within a linear acetonitrile gradient, heteroduplexes that form in PCR samples containing internal sequence variations display reduced column retention time with respect to their homoduplex counterparts. The elution profiles for such samples are distinct from those having a homozygous sequence, making the identification of samples harboring polymorphisms or mutations a straightforward procedure. Only those samples showing abnormal elution profiles are then subjected to sequencing (with or without cloning) to characterize the precise sequence variation(s).
As previously reported,21 experiments with serial dilutions of wild-type and mutant ABL PCR products at different ratios showed that D-HPLC reaches a lower detection limit between 2% and 5% for G250E, E255K, Y253F, H396R, and between 5% and 10% for M351T. These percentages are in agreement with the sensitivity reported in the literature.27-29 Even though it is much more sensitive than direct sequencing and at least as sensitive as subcloning techniques (depending on the number of clones analyzed),21,30 it is currently unclear whether D-HPLC may allow the detection of mutations conferring resistance as early as before the start of imatinib therapy. In our experience, however, mutations were already detectable in all patients before progression to AP/BC, thus allowing for timely adjustments to the overall therapeutic strategy (ie, dose escalation, bone marrow transplantation for eligible patients, or novel tyrosine kinase inhibitors).31-35
In vitro studies have suggested that different mutations confer different degrees of resistance to imatinib.36 Although some mutations (ie, T315I) confer a true resistant phenotype, thereby suggesting withdrawal of imatinib in favor of alternative treatment options, others (ie, M351T) might be overcome by a dose escalation. However, these observations must be confirmed in a clinical setting. Unfortunately, in our study we were unable to address this issue, given that all of the patients herein reported received a standard imatinib dose of 400 mg/d throughout the entire study period.
Patients with and without a missense mutation did not differ regarding sex, median age at the time imatinib therapy was started, and disease history (intolerance/hematologic resistance/cytogenetic resistance to IFN-). On the contrary, median time from diagnosis to the start of imatinib therapy was longer for patients with mutations compared with patients without mutations (4.8 v 2.9 years, respectively; the difference was borderline statistically significant). This finding is in agreement with that reported by Branford et al20 and supports the concept of a gradual accumulation of a pool of BCR-ABL mutants over time, which expands under the selective pressure of imatinib therapy if favored by a lower affinity for the inhibitor.
Few studies have so far dealt with the prognostic impact of ABL point mutations, and contrasting results have been reported. Shah et al10 screened for mutations 13 cytogenetic nonresponders and found that three of four patients with mutations experienced disease progression within 18 months from mutation detection. In contrast, only one of nine patients without a detectable mutation experienced disease progression. Conversely, Hochhaus et al12 did not find any difference in the setting of hematologic resistance between patients with and without a mutation regarding time to progression. Our findings suggest that in late CP CML patients with primary cytogenetic resistance to imatinib, the presence of a point mutation is significantly associated with a greater likelihood of subsequent progression. As many as 40% of late-CP patients fail to reach an MCgR while receiving imatinib therapy,37 and it has been reported that these patients experience disease progression to BC more quickly than those who obtain cytogenetic remission.2 Our results suggest that, within cytogenetic nonresponders, the presence of ABL mutations may identify a subset of patients with particularly poor prognosis in terms of time to progression. This poor prognosis was also confirmed in terms of survival.
Recently, Branford et al20 have shown that in late CP and AP CML patients, a specific subgroup of mutations (ie, those falling within the P-loop) are significantly associated with a particularly poor prognosis in terms of survival. In our series, nine patients showed mutations within the P-loop (G250E, Y253F, Y253H, E255K, E255V). These patients had a significantly worse outcome in terms of time to progression with respect to the remaining patients with mutations (P = .032). Similarly, patients with mutations falling within the P-loop had a significantly shorter survival (P = .045). Therefore, among patients with mutations, mutations mapping within the P-loop seemed to confer a particularly poor outcome. The reasons why P-loop mutations seem to be associated with such an aggressive phenotype currently are unclear. P-loop is a highly conserved region of the kinase domain involved in ATP binding. P-loop mutations such as the Y253F and E255K have demonstrated oncogenic activity in c-ABL.38,39 Moreover, the Y253F and E255K mutants have been shown to induce significantly more tyrosine phosphorylation than wild-type BCR-ABL.7 Interestingly, among several mutated forms of BCR-ABL, the E255K/V, Y253F/H, and G250E showed (together with the T315I, which we did not find in our series) the highest concentration that inhibits 50% values in biochemical and cellular assays.36 Conversely, several mutations falling outside the P-loop (M244V, F311L, E355G, and F359V, among those found in our study) showed the highest concentration that inhibits 50% values near or below the reported trough levels of imatinib in patients treated with 400 mg/d.1,36 These in vitro data are in good (though not perfect) correlation with in vivo observations about clinical outcome: in our series, one of the patients bearing the M244V and the patients bearing the F311L and E355G mutations did not experience disease progression. The second patient bearing the M244V mutation, who experienced disease progression to BC, also had cytogenetic evidence of a second Philadelphia chromosome (data not shown), which certainly contributed to imatinib resistance.
In summary, we found that 45% of patients who did not reach an MCgR at 12 months of therapy with imatinib at the standard dose of 400 mg/d had evidence of a point mutation. Mutations were already detectable by D-HPLC at a median of 3 months from the onset of therapy, and were associated with a greater likelihood of subsequent progression to AP/BC and shorter survival. P-loop mutations were frequent and seemed to confer a particularly poor outcome both in terms of time to progression and in terms of survival. These data suggest that D-HPLC is a valuable tool allowing a timely detection of mutations, which are likely to lead to imatinib resistance. A thorough mutational screening by D-HPLC during the first months of imatinib therapy may play a role in identifying patients with worse prognosis, for whom alternative therapeutic options should be considered.
Appendix
The following members of the ICSG on CML have participated to this study: E. Pogliani and M. Miccolis (Monza); M. Gobbi and M. Miglino (Genova); M. Lazzarino and S. Merante (Pavia); R. Fanin and M. Tiribelli (Udine); D. Russo (Brescia); G. Alimena and E. Montefusco (Roma); G. Rossi and A. Capucci (Brescia); F. Nobile and M. Martino (Reggio Calabria); A. Bacigalupo and F. Frassoni (Genova); B. Rotoli and L. Luciano (Napoli); F. Ferrara and E. Schiavone (Napoli); M. Martelli and A. Tabilio (Perugia); T. Barbui and R. Bassan (Bargamo); V. Rizzoli and L. Mangon (Parma); F. Lauria and M. Bocchia (Siena); E. Volpe and F. Palmieri (Avellino); S. Amadori and A. Cantonetti (Roma); M. Pini (Alessandria); G. Specchia (Bari); A di Tucci (Cagliari); G.L. Scapoli (Ferrara); E. Pungolino (Milano); F. Iuliano (Catanzaro); S. Rupoli (Ancona); P. Guglielmo (Catania); F. Porretto (Palermo); A. Liberati (Perugia); E. Zuffa (Ravenna); M. Cervellera (Taranto); D. Ferrero (Torino); M. Candela (Ancona); C. Bergonzi (Cremona); D. Noli (Nuoro); G. Marini Caracciolo (Palermo); A. Bonati (Parma); F. Parineschi (Pisa); P. Pregno (Torino); A. Ambrosetti (Verona).
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Daniele Alberti, Novartis. For a detailed description of this category, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We thank Vilma Mantovani, Daniela Bastia, and Marinella Cenci (Center for Applied Biomedical Research [CRBA], S. Orsola Hospital, Bologna) for their valuable contribution to D-HPLC analysis. The skillful assistance of Katia Vecchi and Maira Marsili is also acknowledged.
NOTES
Supported by European LeukemiaNet, Cofin 2003 (M.B.), Associazione Italiana Contro le Leucemie-Linfomi e Mieloma, Associazione Italiana per la Ricerca sul Cancro, Fondazione Del Monte di Bologna e Ravenna, Fondo per Gli Investimenti della Ricerca di Base, and Ateneo 60% grants.
No previous article reporting the same set of data has been published.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Druker BJ, Talpaz M, Resta DJ, et al: Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 344:1031-1037, 2001
Kantarjian H, Sawyers C, Hochhaus A, et al: Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N Engl J Med 346:645-652, 2002
Talpaz M, Silver RT, Druker BJ, et al: Imatinib induces durable hematologic and cytogenetic responses in patients with accelerated phase chronic myeloid leukemia: Results of a phase 2 study. Blood 99:1928-1937, 2002
Mauro MJ, O'Dwyer M, Heinrich MC, et al: STI571: A paradigm for cancer therapeutics. J Clin Oncol 20:325-334, 2002
Gorre ME, Mohammed M, Ellwood K, et al: Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 293:876-880, 2001
Branford S, Rudzki Z, Walsh S, et al: High frequency of point mutations clustered within the adenosine triphosphate-binding region of BCR/ABL in patients with chronic myeloid leukemia or Ph-positive acute lymphoblastic leukemia who develop imatinib (STI571) resistance. Blood 99:3472-3475, 2002
Roumiantsev S, Shah NP, Gorre ME, et al: Clinical resistance to the kinase inhibitor STI-571 in chronic myeloid leukemia by mutation of Tyr-253 in the Abl kinase domain P-loop. Proc Natl Acad Sci U S A 99:10700-10705, 2002
von Bubnoff N, Schneller F, Peschel C, et al: BCR-ABL gene mutations in relation to clinical resistance of Philadelphia-chromosome-positive leukaemia to STI571: A prospective study. Lancet 359:487-491, 2002
Roche-Lestienne C, Soenen-Cornu V, Grardel-Duflos N, et al: Several types of mutations of the Abl gene can be found in chronic myeloid leukemia patients resistant to STI571, and they can pre-exist to the onset of treatment. Blood 100:1014-1018, 2002
Shah NP, Nicoll JM, Nagar B, et al: Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor Imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell 2:117-125, 2002
Hofmann WK, Jones LC, Lemp NA, et al: Ph (+) acute lymphoblastic leukemia resistant to the tyrosine kinase inhibitor STI571 has a unique BCR-ABL gene mutation. Blood 99:1860-1862, 2002
Hochhaus A, Kreil S, Corbin AS, et al: Molecular and chromosomal mechanisms of resistance to Imatinib (STI571) therapy. Leukemia 16:2190-2196, 2002
Schindler T, Bornmann W, Pellicena P, et al: Structural mechanism for STI-571 inhibition of Abelson tyrosine kinase. Science 289:1938-1942, 2000
Nagar B, Bornmann WG, Pellicena P, et al: Crystal structures of the kinase domain of c-Abl in complex with the small molecule inhibitors PD173955 and Imatinib (STI-571). Cancer Res 62:4236-4243, 2002
Druker BJ, Tamura S, Buchdunger E, et al: Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 2:561-566, 1996
Gambacorti-Passerini C, le Coutre P, Mologni L, et al: Inhibition of the ABL kinase activity blocks the proliferation of BCR/ABL+ leukemic cells and induces apoptosis. Blood Cells Mol Dis 23:380-394, 1997
Deininger MW, Goldman JM, Lydon N, et al: The tyrosine kinase inhibitor CGP57148B selectively inhibits the growth of BCR-ABL-positive cells. Blood 90:3691-3698, 1997
Bohmer FD, Karagyozov L, Uecker A, et al: A single amino acid exchange inverts susceptibility of related receptor tyrosine kinases for the ATP site inhibitor STI-571. J Biol Chem 278:5148-5155, 2003
Corbin AS, Buchdunger E, Pascal F, et al: Analysis of the structural basis of specificity of inhibition of the Abl kinase by STI571. J Biol Chem 277:32214-32219, 2002
Branford S, Rudzki Z, Walsh S, et al: Detection of BCR-ABL mutations in patients with CML treated with Imatinib is virtually always accompanied by clinical resistance, and mutations in the ATP phosphate-binding loop (P-loop) are associated with a poor prognosis. Blood 102:276-283, 2003
Soverini S, Martinelli G, Amabile M, et al: A denaturing-high performance liquid chromatography (D-HPLC)-based assay for detection of ABL mutations in chronic myeloid leukemia patients resistant to Imatinib. Clin Chem 50:1205-1213, 2004
Rosti G, Martinelli G, Bassi S, et al: Molecular response to Imatinib in late chronic phase chronic myeloid leukemia. Blood 103:2284-2290, 2004
Martinelli G, Ottaviani E, Testoni N, et al: Long-term disease-free acute myeloblastic leukemia with inv(16) is associated with PCR undetectable CBF-beta/MYH11 transcript. Haematologica 85:552-555, 2000
Kreuzer KA, Le Coutre P, Landt O, et al: Preexistence and evolution of imatinib mesylate-resistant clones in chronic myelogenous leukemia detected by a PNA-based PCR clamping technique. Ann Hematol 82:284-289, 2003
Hofmann WK, Komor M, Wassmann B, et al: Presence of the BCR-ABL mutation Glu255Lys prior to STI571 (imatinib) treatment in patients with Ph+ acute lymphoblastic leukemia. Blood 102:659-661, 2003
Xiao W, Oefner P: Denaturing high-performance liquid chromatography: A review. Hum Mutat 17:439-474, 2001
Wolford JK, Blunt D, Ballecer K, et al: High-throughput SNP detection by using DNA pooling and denaturing high performance liquid chromatography (DHPLC). Hum Genet 107:483-487, 2000
Baumer A, Wiedermann U, Hergersberg M, et al: A novel MSP/DHPLC method for the investigation of the methylation status of imprinted genes enables the molecular detection of low cell mosaicisms. Hum Mutat 17:423-430, 2001
Emmerson P, Maynard J, Jones S, et al: Characterizing mutations in samples with low-level mosaicism by collection and analysis of DHPLC fractionated heteroduplexes. Hum Mutat 21:112-115, 2003
Deininger MWN, McGreevey L, Willis S, et al: Detection of ABL kinase domain mutations with denaturing high-performance liquid chromatography. Leukemia 18:864-871, 2004
Shah NP, Tran C, Lee FY, et al: Overriding imatinib resistance with a novel ABL kinase inhibitor. Science 305:399-401, 2004
O'Hare T, Pollock R, Stoffregen EP, et al: Inhibition of wild-type and mutant Bcr-Abl by AP23464, a potent ATP-based oncogenic protein kinase inhibitor: Implications for CML. Blood 104:2532-2539, 2004
Giles F, Kantarjian H, Wassmann B, et al: A phase I/II study of AMN107, a novel aminopyrimidine inhibitor of Bcr-Abl, on a continuous daily dosing schedule in adult patients (pts) with imatinib-resistant advanced phase chronic myeloid leukemia (CML) or relapsed/refractory Philadelphia chromosome (Ph+) acute lymphocytic leukemia (ALL). Blood 104:10a, 2004 (abstr)
Talpaz M, Kantarjian HM, Shah NP, et al: Hematologic and cytogenetic responses in imatinib-resistant accelerated and blast phase chronic myeloid leukemia (CML) patients treated with the dual SRC/ABL kinase inhibitor BMS-354825: Results from a phase I dose escalation study. Blood 104:10a, 2004 (abstr)
Sawyers CL, Shah NP, Kantarjian HM, et al: Hematologic and cytogenetic responses in imatinib-resistant chronic phase chronic myeloid leukemia patients treated with the dual SRC/ABL kinase inhibitor BMS-354825: results from a phase I dose escalation study. Blood 104:4a, 2004 (abstr)
Corbin AS, Rosee PL, Stoffregen EP, et al: Several Bcr-Abl kinase domain mutants associated with imatinib mesylate resistance remain sensitive to imatinib. Blood 101:4611-4614, 2003
Capdeville R, Silberman S: Imatinib: A targeted clinical drug development. Semin Hematol 40:15-20, 2003
Allen P, Weidemann M: An activating mutation in the ATP binding site of the ABL kinase domain. J Biol Chem 271:19585-19591, 1996
Yamamoto M, Kurosu T, Kakihana K, et al: The two major imatinib resistance mutations E255K and T315I enhance the activity of BCR/ABL fusion kinase. Biochem Biophys Res Commun 319:1272-1275, 2004(Simona Soverini, Giovanni)
CEINGE Advanced Biotechnologies and Department of Biochemistry and Medical Biotechnology, University of Naples "Federico II," Naples
Division of Hematology and Internal Medicine, Department of Clinical and Biological Sciences, University of Turin, Turin
Novartis Pharma, Origgio, Italy
ABSTRACT
PURPOSE: Point mutations within the ABL kinase domain of the BCR-ABL gene have been associated with clinical resistance to imatinib mesylate in chronic myeloid leukemia (CML) patients. To shed further light on the frequency, distribution, and prognostic significance of ABL mutations, we retrospectively analyzed a homogeneous cohort of late chronic phase CML patients who showed primary cytogenetic resistance to imatinib.
PATIENTS AND METHODS: Using denaturing high-performance liquid chromatography (D-HPLC) and sequencing, we screened for ABL mutations in a total of 178 bone marrow and/or peripheral blood samples from 40 late chronic phase CML patients homogeneously treated with imatinib 400 mg/d, who did not reach a major cytogenetic response at 12 months.
RESULTS: Mutations were found in 19 of 40 patients (48%). Mutations were already detectable by D-HPLC at a median of 3 months from the onset of therapy. The presence of a missense mutation was significantly associated with a greater likelihood of subsequent progression to accelerated phase/blast crisis (P = .0002) and shorter survival (P = .001). Patients carrying mutations falling within the P-loop seemed to have a particularly poor outcome in terms of time to progression (P = .032) and survival (P = .045).
CONCLUSION: Our results show that, irrespective of the hematologic response, monitoring for emerging mutations in the first months of therapy may play a role in detecting patients with worse prognosis, for whom a revision of the therapeutic strategy should be considered.
INTRODUCTION
The exciting results from trials of imatinib mesylate in patients with chronic myeloid leukemia (CML) have established it as the new standard of care for the chronic phase (CP) and accelerated phase (AP) of the disease.1-4 Imatinib is a potent and selective inhibitor of the BCR-ABL tyrosine kinase, which is known to be deregulated in as many as 95% of CML patients. However, despite high rates of hematologic and cytogenetic responses, primary refractoriness and acquired resistance have been observed in both late CP and AP patients. Point mutations within the ABL kinase domain have been reported as the most frequent mechanism for BCR-ABL reactivation within the leukemic clone.5-12 Structural data suggest that imatinib acts by blocking the BCR-ABL protein into the inactive conformation.13,14 This prevents the transfer of phosphate from adenosine triphosphate (ATP) to substrates and blocks the downstream signal transduction pathways.15-17 Mutations seem to act by disrupting critical contact points between imatinib and BCR-ABL or, more often, by inducing a transition from the inactive to the active state, a conformation to which imatinib is unable to bind.18,19
analyses of clinical samples, the repertoire of mutations found in association with the resistant phenotype has been accruing slowly but inexorably over time, with an overall frequency ranging from 26% to 90% of patients.5-12 Such a difference primarily reflects heterogeneity in patient populations. Most of the studies examined mixed populations of both Philadelphia-positive acute lymphoblastic leukemia and CML patients, the latter at different stages of disease. The criteria for patient selection and the definitions of resistance were also different from one study to another. Most of the studies have focused on patients experiencing disease relapse while receiving imatinib therapy, whereas few data are available to date on the incidence of mutations in patients with primary resistance to imatinib.10,12
Another issue that deserves additional elucidation is the prognostic significance of ABL mutation detection. Shah et al10 screened for mutations in 13 cytogenetic nonresponders and found that three of four patients with mutations experienced disease progression within 18 months from mutation detection. In contrast, only one of nine patients without a detectable mutation experienced disease progression. Hochhaus et al12 examined 43 patients with hematologic relapse and did not find any difference between patients with and without mutations regarding time to progression. Branford et al20 recently showed that in late CP and AP CML patients, a specific subgroup of mutations (ie, those in the ATP phosphate-binding loop [P-loop]) are significantly associated with a poor prognosis in terms of survival.
We recently described a novel denaturing high-performance liquid chromatography (D-HPLC) -based assay for screening of ABL kinase mutations.21 To shed further light on the frequency and prognostic significance of ABL mutations in CML patients resistant to imatinib, we have applied this method to retrospectively analyze a homogeneous cohort of 40 late CP CML patients resistant or refractory to interferon alfa (IFN-) enrolled onto the CML/002/STI571 multicenter clinical trial of the Gruppo Italiano Malattie Ematologiche dell’Adulto (GIMEMA) Working Party on CML (formerly the Italian Cooperative Study Group on CML),22 who showed up-front cytogenetic resistance to imatinib.
Our results confirm that ABL mutations are a relevant mechanism of resistance also in the subset of late-CP patients who are cytogenetically refractory to imatinib. In addition, we found that mutations are significantly associated with a greater likelihood of progression to AP/blast crisis (BC) and death as a result of disease, and that mutations falling within the P-loop confer a particularly poor prognosis.
PATIENTS AND METHODS
Patients and Experimental Design
The study was performed on a total of 178 bone marrow (BM) and/or peripheral blood (PB) samples obtained from 40 patients enrolled onto the CML/002/STI571 multicenter clinical trial of the GIMEMA Working Party on CML.22 The trial enrolled 191 CML patients between June and December 2000. To be eligible, patients were required to be in first CP and to be either intolerant of or resistant to previous IFN- treatment. Patients resistant to IFN- were divided into two groups: hematologic-resistant patients (ie, those who failed to achieve or to maintain a complete hematologic response with IFN-) and cytogenetic-resistant patients (ie, those who failed to achieve or to maintain a complete or partial cytogenetic response with IFN-).2,22 All of the patients provided signed informed consent for participation in the study. Treatment consisted of imatinib 400 mg once daily until disease progression. If hematologic toxicity grade 3 or 4 occurred, the treatment was discontinued until the toxicity had resolved to grade 2 or less. If nonhematologic toxicity grade 2 or 3 occurred, the treatment was discontinued until the toxicity had resolved to less than grade 2, and was resumed at 400 mg if toxicity was grade 2 and at 300 mg if toxicity was grade 3. If grade 4 nonhematologic toxicity occurred, imatinib was permanently withheld.22
The patients who entered onto this molecular study were part of a larger series (n = 56) who showed primary cytogenetic resistance to imatinib therapy (ie, they failed to obtain at least a major cytogenetic response [MCgR] at 12 months). The selection of 40 patients for this study was exclusively based on the availability of samples at our institution. The median time from diagnosis to the start of imatinib therapy was 3.7 years (range, 1.3 to 16.3 years). Sixteen of 40 patients (40%) permanently discontinued imatinib because of disease progression (10 patients) or toxicity (six patients). After imatinib discontinuation, one patient underwent an allogeneic transplantation. The remaining 15 patients were managed with conventional chemotherapy (mainly hydroxyurea) until the time of last contact or death.
ABL mutational screening was done either on the sample(s) collected at month 12 of therapy for those patients who regularly completed the first year of treatment (n = 32), or on the sample(s) collected immediately before imatinib discontinuation for those patients who exited the protocol before month 12 because of disease progression (n = 5) or toxicity (n = 3). Exceptions were patients 35, 36, 37, and 38, who were analyzed at month 6, 6, 9, and 9 of therapy, respectively, because the samples were unavailable at the subsequent time points. Whenever a mutation was detected, its presence was retrospectively traced on the samples collected at the previous time points, when available, back to the pretreatment sample.
Definitions
The cytogenetic response was based on the evaluation of a minimum of 20 marrow cells. MCgR was defined as more than 65% Philadelphia-negative metaphases. Hematologic response was defined as complete (CHR) if the leukocyte count was less than 10 x 109/L, if no immature cells were recorded in the differential count, if the platelet count was less than 450 x 109/L, and if the spleen was not palpable. AP was defined as a percentage of blasts in PB or BM 15% but less than 30%, a percentage of blasts plus promyelocytes in PB or BM 30%, a percentage of peripheral basophils 20%, or thrombocytopenia less than 100 x 109/L unrelated to therapy. BC was defined as 30% blasts in PB or BM.
RNA Extraction and Reverse Transcriptase Polymerase Chain Reaction
Total cellular RNA was extracted from mononuclear cells obtained by Ficoll-Hypaque density gradient centrifugation and resuspended in guanidinium thiocyanate as previously described.23 RNA was spectrophotometrically quantified at 260 nm and its integrity was assessed by electrophoresis on 2% agarose gel. One microgram of total cellular RNA was reverse transcribed to cDNA using 5 μmol/L of random hexamer primers and 200 U of M-MLV reverse transcriptase (GeneAmp RNA PCR Kit, Applied Biosystems, Foster City, CA).
D-HPLC Analysis
For ABL kinase domain amplification, we used a nested polymerase chain reaction (PCR) approach as previously reported.21 All PCR products were analyzed using a D-HPLC Wave 3500HT DNA Fragment Analysis System (Transgenomic Ltd, Crewe Cheshire, United Kingdom). Sample preparation and elution conditions have been described previously.21 One wild-type sample was used as a negative control. Chromatograms from each tested patient were overlaid with the wild-type profile, and samples with extra peak(s) were scored as positive. To ensure that homozygous mutations could not escape D-HPLC detection, for all samples to be studied a mixture of a wild-type PCR product and patient PCR product in a 1:1 ratio was also run. Sensitivity and reliability of the D-HPLC assay were accurately determined. To assess the sensitivity of the D-HPLC assay, which has been reported to be variable in a range of 1% to 10% depending on the sequence and length of the fragments to be analyzed, limiting dilution experiments were performed by mixing known quantities of amplified wild-type and mutant (G250E, Y253F, E255K, T315I, M351T, H396R) ABL PCR products. The percentages of mutant with respect to wild-type were 50:50% –40:60% –30:70% –25:75% –20:80% –17:83% –15:85% –12:88% –10:90% –7:93% –5:95% –2:98% –1:99%. Results indicated that D-HPLC can reach a lower detection limit between 2% and 5% for the following mutations: G250E, Y253F, E255K, T315I, and H396R, and between 5% and 7% for the M351T ABL mutation.21 To rule out the possibility of false-positive and/or false-negative results, 51 BM and/or PB samples from 27 CML patients were analyzed in parallel by D-HPLC and sequencing. Whenever D-HPLC analysis showed an abnormal elution profile, sequence analysis confirmed the presence of a point mutation. Conversely, all of the samples scored as wild-type by D-HPLC did not show evidence of mutations by direct sequencing.21
Direct Sequencing
Direct sequencing was done using the Big Dye Terminator Cycle Sequencing Kit (Applied Biosystems) and an ABI PRISM 377 DNA Analyzer (Applied Biosystems), as previously reported.21 Sequences were compared with the wild-type sequence using BLAST (ABL accession number X16416). Sequence analysis was performed on both strands for each fragment.
Cloning
In three patients for whom direct sequencing failed to detect any point mutation, PCR products were cloned into a pCR2.1-TA vector (TOPO TA Cloning Kit; Invitrogen, Carlsbad, CA) according to the manufacturer's instructions, and 20 independent clones were sequenced.
Statistical Analysis
To test for differences between patients with mutations and patients without mutations, the Mann-Whitney U test was used for continuous variables and the Fisher's exact test or 2 test was used for categoric variables, as appropriate. Curves for survival and time to progression were estimated using the Kaplan-Meier method. The log-rank test was used to test for differences in survival between groups. All analyses were performed using the Statistical Package for the Social Sciences (SPSS software package; SPSS Inc, Chicago, IL).
RESULTS
D-HPLC and Sequence Analyses for ABL Kinase Mutations
In 19 of 40 patients (48%), D-HPLC analysis showed an abnormal elution profile suggesting the presence of one or more sequence variations. Direct sequencing confirmed the presence of a point mutation in all cases. Results of D-HPLC and sequence analyses are listed in Table 1. A total of 13 different point mutations were detected, all but one leading to an amino acid substitution. Nine patients showed mutations (G250E, Y253F, Y253H, E255K, E255V) falling within the nucleotide-binding loop (P-loop). Ten patients showed mutations outside the P-loop, in other regions of the kinase domain (M244V, F311L, F317L, M351T, E355G, F359V, H396R, and a silent mutation at codon 298).
The 19 patients who had a mutation were then subjected to a longitudinal analysis to trace the presence of the mutation in the available samples collected and stored at previous time points, back to the pretreatment sample. Results are reported in Figure 1. Mutations were already detectable at a median of 3 months (range, 1 to 6 months) from the onset of imatinib therapy. In three patients, D-HPLC analysis of the samples collected after 1 (patient 17) and 2 months (patients 5 and 10) from imatinib onset yielded an abnormal elution profile but the presence of the mutation (F359V, Y253F, and E255K, respectively) was revealed only after cloning in two of 20, three of 20, and three of 20 clones, respectively. D-HPLC analysis did not show evidence of mutations before imatinib initiation in any of the patients.
Correlations Between Point Mutations, Baseline Features, and Outcome
Baseline features of patients with mutations versus patients without mutations are listed in Table 2. Mutational status and clinical outcome for each of the 40 patients analyzed are listed in Table 3. As shown in Table 2, no differences emerged between patients with a missense mutation and patients without a missense mutation regarding sex, median age at the time imatinib treatment was started, and disease history (intolerance/hematologic resistance/cytogenetic resistance to IFN-). Median time from diagnosis to the start of imatinib therapy was 4.8 years for patients with missense mutations and 2.9 years for patients without missense mutations; the difference was borderline statistically significant (P = .056).
The CHR rate was similar in the two groups of patients: at 3 and 6 months, the CHR rate was 83% (15 of 18) and 89% (16 of 18), respectively, for patients with missense mutations, compared with 82% (18 of 22) and 90% (20 of 22), respectively, for patients without missense mutations. At the time of mutation detection, all but four patients with missense mutations were in CHR. The median follow-up was 32.5 months (range, 10 to 36 months). As shown in Table 3, 11 of 18 patients with ABL missense mutations experienced disease progression to AP or BC after a median of 9 months (range, 6 to 30 months) from the time mutation was first detected and a median of 12 months from the onset of imatinib therapy (range, 7 to 33 months); seven of them subsequently died as a result of their disease. In contrast, only two patients of 22 without ABL missense mutations experienced disease progression to AP or BC and none of them has since died (Table 3). Curves for time to progression and survival of patients with a mutation versus patients without a mutation are shown in Figures 2 and 3, respectively. Patients with ABL mutations had a significantly shorter time to progression (log-rank P = .0002) and survival (log-rank P = .001).
Among patients with missense mutations, eight of nine patients with mutations falling within the P-loop (G250E, Y253F, Y253H, E255K, E255V) experienced disease progression to AP or BC after a median of 9 months from the time the mutation was first detected (range, 6 to 29 months) and 12 months from the onset of imatinib (range, 10 to 32 months); six of them subsequently died as a result of their disease. In contrast, only three of nine patients with mutations outside the P-loop experienced disease progression to AP or BC, and only one patient died. Curves for time to progression and survival of patients with and without P-loop mutations are shown in Figures 4 and 5, respectively. Patients with P-loop mutations had a significantly shorter time to progression to AP/BC (log-rank P = .032) and survival (log-rank P = .045).
DISCUSSION
The aim of the present study was to examine the frequency, distribution, and prognostic relevance of ABL kinase domain mutations in a homogeneous cohort of late CP CML patients enrolled onto the CML/002/STI571 multicenter clinical trial.22 Taking advantage of a recently validated D-HPLC-based assay for ABL mutational screening,21 we analyzed 40 patients who showed up-front cytogenetic resistance to imatinib, defined as a failure to reach an MCgR at 12 months. Our results show that ABL kinase mutations are a relevant mechanism of resistance to imatinib also in the poorly investigated setting of primary cytogenetic resistance: we detected a total of 13 different point mutations, all but one leading to an amino acid substitution, in 48% of our patients. Nine patients showed missense mutations falling within the nucleotide-binding loop (P-loop; G250E, Y253F, Y253H, E255K, E255V). Ten patients showed missense mutations in other regions of the kinase domain (M244V, F311L, F317L, M351T, E355G, F359V, H396R). All of these mutations have been described previously.5-12,20,24 In a patient from our series we also found a single amino acid substitution leading to a silent mutation at codon 298. The mutated clone coexisted with the wild-type one at an approximate relative ratio of 50% to 50%. The patient was investigated further to trace the presence of the mutation in the available samples collected and stored at previous time points, back to the pretreatment sample (Fig 1). The nucleotide change was detectable at 6 months and in all the samples collected thereafter, but not in those collected before the onset of therapy and at 1 and 2 months. These observations raise a question regarding the biologic significance of this amino acid substitution. Because mechanisms of resistance other than mutations exist, it may be speculated that the mutated BCR-ABL allele, not conferring per se any selective advantage, originated in a subclone bearing other abnormalities actually responsible for the selection and/or expansion of the subclone itself and for the resistant phenotype.
The same longitudinal, retrospective analysis was performed on all of the other patients bearing mutations (Fig 1) to determine how early during treatment D-HPLC could reveal the mutation, thus evaluating whether our screening method could be predictive of an emerging resistance. Overall, mutations were already detectable by D-HPLC at a median of 3 months (range, 1 to 6 months) from the onset of therapy. All patients but four had sustained hematologic response at the time of mutation detection. Bearing in mind that all the patients herein reported did not reach an MCgR at 12 months, our results indicate that as many as 50% of patients who failed to achieve cytogenetic control of the disease in the first 6 months of therapy with imatinib already had evidence of a point mutation by D-HPLC analysis. These data would suggest that, irrespective of the hematologic response, regular monitoring for emerging mutations in the first months of therapy might help to identify patients for whom a revision of the therapeutic strategy should be taken into account.
Despite allowing the identification of resistance-associated mutations as early as 1 month after the onset of imatinib, D-HPLC screening did not detect mutations in any of the pretreatment samples analyzed. It has been reported recently by some authors that, in some CML and Philadelphia-positive acute lymphoblastic leukemia patients, specific mutations (E255K, F311L, T315I) could be found in a small proportion of leukemic cells before the onset of imatinib treatment.9,24,25 The reason why other studies12,20 failed to detect mutations in pretreatment samples might be a lack of sufficient sensitivity of the techniques used for mutational screening. D-HPLC is a reverse-phase ion pair HPLC specifically developed for detection of DNA sequence variations such as point mutations, small insertions, and deletions.26 Under conditions of partial heat denaturation within a linear acetonitrile gradient, heteroduplexes that form in PCR samples containing internal sequence variations display reduced column retention time with respect to their homoduplex counterparts. The elution profiles for such samples are distinct from those having a homozygous sequence, making the identification of samples harboring polymorphisms or mutations a straightforward procedure. Only those samples showing abnormal elution profiles are then subjected to sequencing (with or without cloning) to characterize the precise sequence variation(s).
As previously reported,21 experiments with serial dilutions of wild-type and mutant ABL PCR products at different ratios showed that D-HPLC reaches a lower detection limit between 2% and 5% for G250E, E255K, Y253F, H396R, and between 5% and 10% for M351T. These percentages are in agreement with the sensitivity reported in the literature.27-29 Even though it is much more sensitive than direct sequencing and at least as sensitive as subcloning techniques (depending on the number of clones analyzed),21,30 it is currently unclear whether D-HPLC may allow the detection of mutations conferring resistance as early as before the start of imatinib therapy. In our experience, however, mutations were already detectable in all patients before progression to AP/BC, thus allowing for timely adjustments to the overall therapeutic strategy (ie, dose escalation, bone marrow transplantation for eligible patients, or novel tyrosine kinase inhibitors).31-35
In vitro studies have suggested that different mutations confer different degrees of resistance to imatinib.36 Although some mutations (ie, T315I) confer a true resistant phenotype, thereby suggesting withdrawal of imatinib in favor of alternative treatment options, others (ie, M351T) might be overcome by a dose escalation. However, these observations must be confirmed in a clinical setting. Unfortunately, in our study we were unable to address this issue, given that all of the patients herein reported received a standard imatinib dose of 400 mg/d throughout the entire study period.
Patients with and without a missense mutation did not differ regarding sex, median age at the time imatinib therapy was started, and disease history (intolerance/hematologic resistance/cytogenetic resistance to IFN-). On the contrary, median time from diagnosis to the start of imatinib therapy was longer for patients with mutations compared with patients without mutations (4.8 v 2.9 years, respectively; the difference was borderline statistically significant). This finding is in agreement with that reported by Branford et al20 and supports the concept of a gradual accumulation of a pool of BCR-ABL mutants over time, which expands under the selective pressure of imatinib therapy if favored by a lower affinity for the inhibitor.
Few studies have so far dealt with the prognostic impact of ABL point mutations, and contrasting results have been reported. Shah et al10 screened for mutations 13 cytogenetic nonresponders and found that three of four patients with mutations experienced disease progression within 18 months from mutation detection. In contrast, only one of nine patients without a detectable mutation experienced disease progression. Conversely, Hochhaus et al12 did not find any difference in the setting of hematologic resistance between patients with and without a mutation regarding time to progression. Our findings suggest that in late CP CML patients with primary cytogenetic resistance to imatinib, the presence of a point mutation is significantly associated with a greater likelihood of subsequent progression. As many as 40% of late-CP patients fail to reach an MCgR while receiving imatinib therapy,37 and it has been reported that these patients experience disease progression to BC more quickly than those who obtain cytogenetic remission.2 Our results suggest that, within cytogenetic nonresponders, the presence of ABL mutations may identify a subset of patients with particularly poor prognosis in terms of time to progression. This poor prognosis was also confirmed in terms of survival.
Recently, Branford et al20 have shown that in late CP and AP CML patients, a specific subgroup of mutations (ie, those falling within the P-loop) are significantly associated with a particularly poor prognosis in terms of survival. In our series, nine patients showed mutations within the P-loop (G250E, Y253F, Y253H, E255K, E255V). These patients had a significantly worse outcome in terms of time to progression with respect to the remaining patients with mutations (P = .032). Similarly, patients with mutations falling within the P-loop had a significantly shorter survival (P = .045). Therefore, among patients with mutations, mutations mapping within the P-loop seemed to confer a particularly poor outcome. The reasons why P-loop mutations seem to be associated with such an aggressive phenotype currently are unclear. P-loop is a highly conserved region of the kinase domain involved in ATP binding. P-loop mutations such as the Y253F and E255K have demonstrated oncogenic activity in c-ABL.38,39 Moreover, the Y253F and E255K mutants have been shown to induce significantly more tyrosine phosphorylation than wild-type BCR-ABL.7 Interestingly, among several mutated forms of BCR-ABL, the E255K/V, Y253F/H, and G250E showed (together with the T315I, which we did not find in our series) the highest concentration that inhibits 50% values in biochemical and cellular assays.36 Conversely, several mutations falling outside the P-loop (M244V, F311L, E355G, and F359V, among those found in our study) showed the highest concentration that inhibits 50% values near or below the reported trough levels of imatinib in patients treated with 400 mg/d.1,36 These in vitro data are in good (though not perfect) correlation with in vivo observations about clinical outcome: in our series, one of the patients bearing the M244V and the patients bearing the F311L and E355G mutations did not experience disease progression. The second patient bearing the M244V mutation, who experienced disease progression to BC, also had cytogenetic evidence of a second Philadelphia chromosome (data not shown), which certainly contributed to imatinib resistance.
In summary, we found that 45% of patients who did not reach an MCgR at 12 months of therapy with imatinib at the standard dose of 400 mg/d had evidence of a point mutation. Mutations were already detectable by D-HPLC at a median of 3 months from the onset of therapy, and were associated with a greater likelihood of subsequent progression to AP/BC and shorter survival. P-loop mutations were frequent and seemed to confer a particularly poor outcome both in terms of time to progression and in terms of survival. These data suggest that D-HPLC is a valuable tool allowing a timely detection of mutations, which are likely to lead to imatinib resistance. A thorough mutational screening by D-HPLC during the first months of imatinib therapy may play a role in identifying patients with worse prognosis, for whom alternative therapeutic options should be considered.
Appendix
The following members of the ICSG on CML have participated to this study: E. Pogliani and M. Miccolis (Monza); M. Gobbi and M. Miglino (Genova); M. Lazzarino and S. Merante (Pavia); R. Fanin and M. Tiribelli (Udine); D. Russo (Brescia); G. Alimena and E. Montefusco (Roma); G. Rossi and A. Capucci (Brescia); F. Nobile and M. Martino (Reggio Calabria); A. Bacigalupo and F. Frassoni (Genova); B. Rotoli and L. Luciano (Napoli); F. Ferrara and E. Schiavone (Napoli); M. Martelli and A. Tabilio (Perugia); T. Barbui and R. Bassan (Bargamo); V. Rizzoli and L. Mangon (Parma); F. Lauria and M. Bocchia (Siena); E. Volpe and F. Palmieri (Avellino); S. Amadori and A. Cantonetti (Roma); M. Pini (Alessandria); G. Specchia (Bari); A di Tucci (Cagliari); G.L. Scapoli (Ferrara); E. Pungolino (Milano); F. Iuliano (Catanzaro); S. Rupoli (Ancona); P. Guglielmo (Catania); F. Porretto (Palermo); A. Liberati (Perugia); E. Zuffa (Ravenna); M. Cervellera (Taranto); D. Ferrero (Torino); M. Candela (Ancona); C. Bergonzi (Cremona); D. Noli (Nuoro); G. Marini Caracciolo (Palermo); A. Bonati (Parma); F. Parineschi (Pisa); P. Pregno (Torino); A. Ambrosetti (Verona).
Authors' Disclosures of Potential Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Daniele Alberti, Novartis. For a detailed description of this category, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Acknowledgment
We thank Vilma Mantovani, Daniela Bastia, and Marinella Cenci (Center for Applied Biomedical Research [CRBA], S. Orsola Hospital, Bologna) for their valuable contribution to D-HPLC analysis. The skillful assistance of Katia Vecchi and Maira Marsili is also acknowledged.
NOTES
Supported by European LeukemiaNet, Cofin 2003 (M.B.), Associazione Italiana Contro le Leucemie-Linfomi e Mieloma, Associazione Italiana per la Ricerca sul Cancro, Fondazione Del Monte di Bologna e Ravenna, Fondo per Gli Investimenti della Ricerca di Base, and Ateneo 60% grants.
No previous article reporting the same set of data has been published.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Druker BJ, Talpaz M, Resta DJ, et al: Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 344:1031-1037, 2001
Kantarjian H, Sawyers C, Hochhaus A, et al: Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N Engl J Med 346:645-652, 2002
Talpaz M, Silver RT, Druker BJ, et al: Imatinib induces durable hematologic and cytogenetic responses in patients with accelerated phase chronic myeloid leukemia: Results of a phase 2 study. Blood 99:1928-1937, 2002
Mauro MJ, O'Dwyer M, Heinrich MC, et al: STI571: A paradigm for cancer therapeutics. J Clin Oncol 20:325-334, 2002
Gorre ME, Mohammed M, Ellwood K, et al: Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 293:876-880, 2001
Branford S, Rudzki Z, Walsh S, et al: High frequency of point mutations clustered within the adenosine triphosphate-binding region of BCR/ABL in patients with chronic myeloid leukemia or Ph-positive acute lymphoblastic leukemia who develop imatinib (STI571) resistance. Blood 99:3472-3475, 2002
Roumiantsev S, Shah NP, Gorre ME, et al: Clinical resistance to the kinase inhibitor STI-571 in chronic myeloid leukemia by mutation of Tyr-253 in the Abl kinase domain P-loop. Proc Natl Acad Sci U S A 99:10700-10705, 2002
von Bubnoff N, Schneller F, Peschel C, et al: BCR-ABL gene mutations in relation to clinical resistance of Philadelphia-chromosome-positive leukaemia to STI571: A prospective study. Lancet 359:487-491, 2002
Roche-Lestienne C, Soenen-Cornu V, Grardel-Duflos N, et al: Several types of mutations of the Abl gene can be found in chronic myeloid leukemia patients resistant to STI571, and they can pre-exist to the onset of treatment. Blood 100:1014-1018, 2002
Shah NP, Nicoll JM, Nagar B, et al: Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor Imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell 2:117-125, 2002
Hofmann WK, Jones LC, Lemp NA, et al: Ph (+) acute lymphoblastic leukemia resistant to the tyrosine kinase inhibitor STI571 has a unique BCR-ABL gene mutation. Blood 99:1860-1862, 2002
Hochhaus A, Kreil S, Corbin AS, et al: Molecular and chromosomal mechanisms of resistance to Imatinib (STI571) therapy. Leukemia 16:2190-2196, 2002
Schindler T, Bornmann W, Pellicena P, et al: Structural mechanism for STI-571 inhibition of Abelson tyrosine kinase. Science 289:1938-1942, 2000
Nagar B, Bornmann WG, Pellicena P, et al: Crystal structures of the kinase domain of c-Abl in complex with the small molecule inhibitors PD173955 and Imatinib (STI-571). Cancer Res 62:4236-4243, 2002
Druker BJ, Tamura S, Buchdunger E, et al: Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 2:561-566, 1996
Gambacorti-Passerini C, le Coutre P, Mologni L, et al: Inhibition of the ABL kinase activity blocks the proliferation of BCR/ABL+ leukemic cells and induces apoptosis. Blood Cells Mol Dis 23:380-394, 1997
Deininger MW, Goldman JM, Lydon N, et al: The tyrosine kinase inhibitor CGP57148B selectively inhibits the growth of BCR-ABL-positive cells. Blood 90:3691-3698, 1997
Bohmer FD, Karagyozov L, Uecker A, et al: A single amino acid exchange inverts susceptibility of related receptor tyrosine kinases for the ATP site inhibitor STI-571. J Biol Chem 278:5148-5155, 2003
Corbin AS, Buchdunger E, Pascal F, et al: Analysis of the structural basis of specificity of inhibition of the Abl kinase by STI571. J Biol Chem 277:32214-32219, 2002
Branford S, Rudzki Z, Walsh S, et al: Detection of BCR-ABL mutations in patients with CML treated with Imatinib is virtually always accompanied by clinical resistance, and mutations in the ATP phosphate-binding loop (P-loop) are associated with a poor prognosis. Blood 102:276-283, 2003
Soverini S, Martinelli G, Amabile M, et al: A denaturing-high performance liquid chromatography (D-HPLC)-based assay for detection of ABL mutations in chronic myeloid leukemia patients resistant to Imatinib. Clin Chem 50:1205-1213, 2004
Rosti G, Martinelli G, Bassi S, et al: Molecular response to Imatinib in late chronic phase chronic myeloid leukemia. Blood 103:2284-2290, 2004
Martinelli G, Ottaviani E, Testoni N, et al: Long-term disease-free acute myeloblastic leukemia with inv(16) is associated with PCR undetectable CBF-beta/MYH11 transcript. Haematologica 85:552-555, 2000
Kreuzer KA, Le Coutre P, Landt O, et al: Preexistence and evolution of imatinib mesylate-resistant clones in chronic myelogenous leukemia detected by a PNA-based PCR clamping technique. Ann Hematol 82:284-289, 2003
Hofmann WK, Komor M, Wassmann B, et al: Presence of the BCR-ABL mutation Glu255Lys prior to STI571 (imatinib) treatment in patients with Ph+ acute lymphoblastic leukemia. Blood 102:659-661, 2003
Xiao W, Oefner P: Denaturing high-performance liquid chromatography: A review. Hum Mutat 17:439-474, 2001
Wolford JK, Blunt D, Ballecer K, et al: High-throughput SNP detection by using DNA pooling and denaturing high performance liquid chromatography (DHPLC). Hum Genet 107:483-487, 2000
Baumer A, Wiedermann U, Hergersberg M, et al: A novel MSP/DHPLC method for the investigation of the methylation status of imprinted genes enables the molecular detection of low cell mosaicisms. Hum Mutat 17:423-430, 2001
Emmerson P, Maynard J, Jones S, et al: Characterizing mutations in samples with low-level mosaicism by collection and analysis of DHPLC fractionated heteroduplexes. Hum Mutat 21:112-115, 2003
Deininger MWN, McGreevey L, Willis S, et al: Detection of ABL kinase domain mutations with denaturing high-performance liquid chromatography. Leukemia 18:864-871, 2004
Shah NP, Tran C, Lee FY, et al: Overriding imatinib resistance with a novel ABL kinase inhibitor. Science 305:399-401, 2004
O'Hare T, Pollock R, Stoffregen EP, et al: Inhibition of wild-type and mutant Bcr-Abl by AP23464, a potent ATP-based oncogenic protein kinase inhibitor: Implications for CML. Blood 104:2532-2539, 2004
Giles F, Kantarjian H, Wassmann B, et al: A phase I/II study of AMN107, a novel aminopyrimidine inhibitor of Bcr-Abl, on a continuous daily dosing schedule in adult patients (pts) with imatinib-resistant advanced phase chronic myeloid leukemia (CML) or relapsed/refractory Philadelphia chromosome (Ph+) acute lymphocytic leukemia (ALL). Blood 104:10a, 2004 (abstr)
Talpaz M, Kantarjian HM, Shah NP, et al: Hematologic and cytogenetic responses in imatinib-resistant accelerated and blast phase chronic myeloid leukemia (CML) patients treated with the dual SRC/ABL kinase inhibitor BMS-354825: Results from a phase I dose escalation study. Blood 104:10a, 2004 (abstr)
Sawyers CL, Shah NP, Kantarjian HM, et al: Hematologic and cytogenetic responses in imatinib-resistant chronic phase chronic myeloid leukemia patients treated with the dual SRC/ABL kinase inhibitor BMS-354825: results from a phase I dose escalation study. Blood 104:4a, 2004 (abstr)
Corbin AS, Rosee PL, Stoffregen EP, et al: Several Bcr-Abl kinase domain mutants associated with imatinib mesylate resistance remain sensitive to imatinib. Blood 101:4611-4614, 2003
Capdeville R, Silberman S: Imatinib: A targeted clinical drug development. Semin Hematol 40:15-20, 2003
Allen P, Weidemann M: An activating mutation in the ATP binding site of the ABL kinase domain. J Biol Chem 271:19585-19591, 1996
Yamamoto M, Kurosu T, Kakihana K, et al: The two major imatinib resistance mutations E255K and T315I enhance the activity of BCR/ABL fusion kinase. Biochem Biophys Res Commun 319:1272-1275, 2004(Simona Soverini, Giovanni)