演題詳細

教育講演 / Educational Lecture

【E】教育講演21 (Educational Lecture 21) : Topics

print

日程
2013年10月11日(金)
時間
13:55 - 14:25
会場
第7会場 / Room No.7 (ロイトン札幌 2F リージェント)
座長・司会
大屋敷 一馬 (Kazuma Ohyashiki):1
1:Department of Hematology, Tokyo Medical University, Japan
 
前へ戻る

A novel BIM deletion polymorphism: implications and lessons for cancer targeted therapies

演題番号 : EL-21

Ong Sin Tiong:1

1:Program in Cancer & Stem Cell Biology, Duke - Nus Graduate Medical School Singapore, Singapore

 

Introduction

Molecularly targeted therapy has revolutionized cancer therapy, and resulted in consistently high response rates, often associated with improvements in progression free survival (PFS) as well as improved overall survival (OS)1). These advances have benefited many patients suffering from cancers previously considered to be highly therapy resistant2, 3). Such advances are based on the ability to identify the molecular drivers of each cancer, and specifically target those drivers pharmacologically.
With over a decade of experience in many such cancer models, we have begun to appreciate that a major issue facing the field is the degree of response heterogeneity observed across different cancers even when they molecularly defined and targeted. This heterogeneity encompasses significant variations in depth of response to initial therapy, duration of response, and OS. A major challenge will be to understand the basis for tumor and patient heterogeneity, and to overcome suboptimal responses so that all individuals with cancer can enjoy optimal responses. Increased knowledge in this area will lead to improved biomarkers for predicting resistance, and disease transformation. Further, if these biomarkers are functionally important, understanding their mechanism of action may also provide novel therapeutic approaches to prevent and/or treat clinical resistance. Toward this end, we outline our recent discovery of a common polymorphism in the BIM gene that accounts for heterogeneous responses in at least two human cancers, chronic myeloid leukemia (CML) and epidermal growth factor receptor-mutated non-small-cell lung cancer (EGFR NSCLC), and the lessons we have learned from it4).

The BIM deletion polymorphism and intrinsic resistance to targeted therapies

To identify structural variations (SVs) that were associated with tyrosine kinase inhibitor (TKI) resistance in CML patients, we used DNA sequencing of paired-end ditags (DNA-PET) to sequence and compare the genomes of three TKI-resistant patients with those of two TKI-sensitive patients. From the list of SVs that occurred exclusively in the TKI-resistant patients, one candidate stood out. This SV was a 2.9 kb deletion that occurred in intron 2 of the BIM (otherwise known as BCL2L11) gene. This was of tremendous interest to us given the well-known role of BIM in mediating TKI-induced apoptosis in a number of cancers 5). Targeted sequencing of the BIM locus revealed that the same deletion was found in all three samples, suggesting germline origin. This was confirmed when we screened a large cohort of normal individuals and found that it occurred at a frequency of 12.3% in East Asian populations, but was absent in African or European populations. Clinically, a retrospective analysis of an East Asian TKI-treated CML cohort revealed an increased risk for resistant disease as defined by European Leukemia Net criteria (OR=2.94, p=0.02, 95% CI=1.17─7.43) in patients with the polymorphism.
We went on to show that the deletion results in a switch in splicing patterns to favor the production of exon 3-containing BIM isoforms. Because exon 3 and 4 are spliced in a mutually exclusive fashion, exon 3-containing isoforms lack the BH3 domain that is found in exon 4 and is required for pro-apoptotic function. To prove that the polymorphism was sufficient to confer resistance to TKIs, we introduced it into the K562 cell line using zinc finger nuclease technology. We demonstrated that the increase in exon 3 to exon 4 ratio was recapitulated in the deletion-containing clones and that apoptosis was attenuated in a polymorphism dose-dependent manner. Using the same approach, we validated our in vitro findings in another kinase-driven cancer, EGFR NSCLC, and showed that patients with EGFR NSCLC had inferior progression-free survival (PFS) of 6.6 vs 11.9 months when compared to patients without the deletion polymorphism. Notably, we were able to overcome resistance mediated by the polymorphism by using ABT-737, a BH3 mimetic6), to pharmacologically restore BIM function in our genome-edited lines as well as primary samples with the deletion.

Assessing causes of response heterogeneity in cancer targeted therapies

Response heterogeneity is a broad term, and can include the quality of the initial response to targeted therapy, PFS, and OS. In hematologic malignancies, including CML, the depth of response is routinely measured by standard quantitative laboratory tests (e.g. quantitative RT-PCR for BCR-ABL1 transcripts), which serve as a useful and objective measure of initial response. In solid tumors, molecularly-based quantitative measures are more challenging to assess, and in practical terms responses are usually limited to measuring best tumor responses by imaging modalities. Accordingly, other surrogate measures of response are frequently used such as PFS and OS. However, OS especially can be subject to confounding factors since progressing patients are usually treated with alternative therapies including cytotoxic chemotherapy and radiotherapy. Thus, when response heterogeneity is assessed it should be specific to the clinical measure being assessed. In the case of EGFR NSCLC, one possible objective clinical measure is the best response on CT scans.
In the broadest sense, the genetic causes for response heterogeneity may be classified as either acquired and somatic, or inherited and germline. Here, it should be appreciated that germline variants may also influence the likelihood of the acquisition of acquired mutations that confer resistance, as discussed in more detail below. Examples of acquired mutations are those that directly mediate resistance to targeted therapies by interfering with the binding of the drug to the active site of the target enzyme. These mutations are thought to arise stochastically in a tumor cell subclone, and as might be expected, have been described to pre-exist the use of the targeted drug7). Examples of germline variants that directly mediate resistance to targeted therapies are much rarer. More commonly, these variants have an indirect effect, and usually affect pharmacologic parameters8). These include polymorphisms that either modulate drug metabolism or drug import/export, and thereby cause resistance by modulating intracellular drug concentrations. We also consider the situation where germline variants may enhance the likelihood of a tumor acquiring secondary mutations. In the case of the BIM deletion polymorphism, it is possible that suboptimal doses of TKIs may allow subclones of tumor cells to survive, proliferate, and accumulate secondary mutations that themselves confer TKI resistance.
Recent reviews have classified clinical resistance to targeted therapies as being primary or secondary9). The latter are defined as patients who fail to achieve any response to initial therapy, while the former are those who have an initial response but who subsequently relapse or progress. Frequently, the causes of primary and secondary clinical resistance are attributed to germline and acquired mutations respectively. However, it should be appreciated that this segregation assumes that germline mutations confer absolute drug resistance, which may not be the case, especially since tumor populations are known to be genetically heterogeneous10). This concept will be discussed further below.
Study design is also critical to the efficient discovery of genetic variants mediating response heterogeneity8). First of all, such studies should be sufficiently powered to detect variants of clinical significance and be independently validated. Secondly, when using genome-wide approaches to identify variants associated with response heterogeneity, it is important to consider the population being studied to enable efficient discovery. As far as possible, the study should include homogenous populations, and in the ideal situation, use a single molecularly defined entity, e.g. EGFR-mutated non-small cell lung cancer, treated with a single agent. Further, if the study is to identify resistance to targeted therapies, then those populations should have been treated only with that agent, otherwise confounding factors will likely be introduced since resistance and/or sensitivity to different therapeutic agents is likely to be multifactorial.

Discovery of the BIM deletion polymorphism

The impetus for the work that led to the discovery of the BIM deletion polymorphism was to uncover mechanisms that mediated intrinsic resistance to ABL1 kinase inhibitors in CML. Prior work had suggested that a significant proportion of individuals with clinical resistance may not harbor mutations in BCR-ABL1, particularly those in early stage disease11). Further, studies in such patients indicated that downstream signalling could still be inhibited by TKIs, yet they experienced suboptimal responses11). Together, these observations suggested that BCR-ABL1-independent factors could be underlying resistance. Accordingly, when selecting CML samples for genomic studies, we deliberately chose material from patients who had overt clinical TKI resistance but did not harbor BCR-ABL1 kinase domain mutations.
A second feature of our workflow was to include polymorphic variants as genes of interest, particularly those that were enriched in the study population, i.e. enriched in patients with resistance compared to those without. In many cancer genome-wide studies, polymorphic variants are usually filtered out to increase the chances of identifying functional variants (such as driver or resistance-conferring mutations). Such approaches would have discarded the BIM deletion polymorphism.
A third feature of our work was combining the pipeline of genetic variants of interest with prompt validation in CML samples segregated according to the phenotype of interest. Thus, because the deletion was almost 3kb in size, PCR primers could be designed to efficiently detect deletion-containing alleles in a single reaction, without necessarily sequencing the PCR products. When these data showed that the deletion was enriched in resistant populations, functional effects of the deletion polymorphism were then investigated at the mechanistic level at the bench.
Finally, two other important factors in the process of discovery should be mentioned. The first was the role played by several groups in clearly and convincingly demonstrating the critical importance of BIM expression in determining sensitivity to TKIs in CML12, 13), and the second was the availability of several East Asian CML cell lines, one of which (KCL22) was found to harbor the BIM deletion polymorphism and to be intrinsically resistant to TKIs14). Here, it should be noted that the latter are almost exclusively Japanese in origin, and we owe a debt of gratitude to investigators in the past for their contribution to our studies.

Lessons from the discovery of the BIM deletion

One of the most important lessons we learnt from the discovery of the BIM deletion polymorphism is the potential for relatively common polymorphisms to contribute significantly to clinical outcome in cancer therapies. This is likely to due to the critical importance that BIM plays in mediating cancer cell death. Indeed, cancer cells (as well as normal cells) are exquisitely sensitive to BIM levels, such that gene dosage effects are clearly seen in animal models of cancer15). However, the BIM deletion polymorphism is not a perfect predictor of resistance, since the odds ratios of an individual with the deletion polymorphism being resistant, while robust and clinically meaningful, were in the range of 2-3-fold. Thus, the deletion polymorphism likely works in concert with other factors, yet to be defined, to determine the ultimate outcome of resistance vs sensitivity.
That the BIM deletion polymorphism is restricted to persons of East Asian ancestry is also of interest. While some reports have indicated that TKI response rates in CML may be inferior in East Asia compared to the West16), response rates for TKIs in other cancers including EGFR TKIs in EGFR NSCLC are not clearly inferior in the East compared to the West17). This suggests that different ethnicities may harbor polymorphic variants of at least equivalent effect to that of the BIM deletion. These may of course reside in the same gene, or affect other genes in the same or similar pathways. Further, these polymorphic variants may not necessarily occur in genic regions, but in gene regulatory regions such as those described by the ENCODE project18).
While the discovery was made in CML, the biology and clinical significance was extended to EGFR NSCLC. Here, we observed that the deletion mediated intrinsic resistance to EGFR inhibitors in cell lines, and that the deletion shortened the PFS from 11.9 to 6.6 months in patients without and with the deletion polymorphism respectively. Two other groups have since published data regarding the BIM deletion polymorphism in EGFR NSCLC. The first was by Dr. Seiji Yano’s group from Kanazawa University19). These investigators discovered a novel EGFR NSCLC cell line with the BIM deletion polymorphism, and demonstrated that it had the expected phenotype comprising intrinsic resistance to EGFR TKIs, increased splicing of BIM exon 3 at the expense of exon 4, and decreased expression of BH3-containing BIM isoforms. More importantly, they made the intriguing finding that the histone deacetylase inhibitor, vorinostat, is able to restore the aberrant splicing of BIM to favor exon 4 over exon 3, resulting in expression of BH3-containing BIM isoforms, and resensitization to EGFR TKIs. As stated by the authors, the exact mechanism by which vorinostat alters BIM splicing remains to be elucidated, although they noted that effects of vorinostat on mRNA splicing have previously been described20). In addition to data supporting our initial discovery, we also take note of a recent publication by a group from Korea with contrary results21). This group performed a retrospective analysis of 197 patients with EGFR NSCLC and concluded that several factors, including the BIM deletion polymorphism, treatment type, EGFR genotype, or smoking did not predict inferior PFS. Ideally, studies to determine the role of the BIM deletion polymorphism should be prospective and include large cohorts of patients, and indeed such a study is under way at the National Cancer Centre in Singapore.
With regard to CML, a small study by Katagiri et al. from Tokyo Medical University has suggested that the BIM deletion polymorphism might be used as a criterion for discontinuation of imatinib in CML22). These authors determined the incidence of the BIM deletion polymorphism among patients with CML treated with TKIs, and who had achieved a complete molecular remission (CMR). Interestingly, out of 5 patients who were able to remain in CMR 12 months after stopping TKI therapy, all 5 were negative for the deletion. In contrast, out of 8 patients who achieved CMR and had disease relapse within 12 months of TKI cessation, 3 had the deletion polymorphism. While the data is not conclusive (given the small number of patients), it is nevertheless very intriguing, and led the authors to suggest that the deletion polymorphism may be influencing relapse rates via affecting the function of T and/or NK cells, which in turn are responsible for immune surveillance. Given the known effects of BIM haploinsufficiency on immune cell function23~25) and the documented anti-leukemia cell effect of T and NK cells26), this idea certainly offers a plausible explanation for their data. In this respect, it would be interesting to determine if rates of relapse among allografted CML patients might be different depending on the BIM deletion status of the donor. Additionally, alternative explanations should also be considered for the findings of Katagiri et al. Given that patients with CML are closely monitored for benchmark responses (usually by measuring the proportion of Philadelphia chromosome positive cells in blood and/or bone marrow, as well as by quantitative RT-PCR for BCR-ABL1 transcripts), it could be possible that those individuals with suboptimal responses may have had their TKI doses increased or their TKI changed to more potent second generation drugs. Furthermore, it should be highlighted that the BIM deletion polymorphism confers relative and not absolute TKI resistance, and as such, the increases in drug dose and/or potency would be expected to result in increased responses. Because CML progenitors are thought to act as reservoirs for relapse27), another possible explanation could be that the deletion polymorphism modulates leukemia stem cell survival and/or maintenance. This possibility is consistent with the inability of clinical tests to reliably detect or quantify residual CML LSCs in patients28), and the recently described reliance of CML LSCs on BCL2 family members for maintenance29).

More work to be done

In addition to CML and EGFR NSCLC, several other human cancers have been documented to be critically dependent on BIM for drug sensitivity. These include c-KIT-driven gastrointestinal tumors (imatinib resistance), JAK2-mutated myeloproliferative disorders (JAK2 inhibitor resistance), and acute lymphoblastic leukemia (ALL)(steroid resistance)30~32). It would therefore be important to determine if patients with the BIM deletion polymorphism do indeed experience inferior responses to the above agents, and if so, could they be resensitized using drugs that restore BIM splicing patterns or replace BIM function. In performing such studies, it should also be noted that concurrent therapy, including cytotoxic agents, may introduce confounding factors. For example, in ALL, two agents which are commonly used in induction therapy are vincristine and L-asparaginase, and both have been described to kill cancer cells by modulating levels of BCL-2 family members33, 34). Thus, any negative clinical effects of the BIM deletion polymorphism regarding steroid responsiveness may be counteracted by the ability of cytotoxics to tip the overall apoptotic threshold toward cell death.
Because several agents have now been found to reverse TKI resistance conferred by the BIM deletion polymorphism, including BH3 mimetic drugs and HDAC inhibitors4, 35), it is now important to test these combinations in patients. It remains to be seen whether or not such studies will include patients without the BIM deletion polymorphism, since in vitro work suggests that patients with and without the deletion may both benefit from BH3 mimetic/TKI combinations.
Finally, two other questions have been raised by our discovery. The first relates to the high frequency of the deletion polymorphism is in East Asians, and its non-existence in African or Caucasian individuals. Given the known role of BIM in immune cell function, could the deletion confer a survival advantage against environmental pathogens? Second, given the established role of BIM as a tumor suppressor, could it be cooperating with acquired EGFR mutations to transform lung epithelium, and thus explain the increased incidence of EGFR NSCLC in East Asia compared to the West?
In summary, the discovery of the BIM deletion polymorphism has helped us to understand how germline variants affecting key genes regulating cell survival and death can have profound effects in the cancer clinic. The discovery has also explained in part why heterogeneous responses to targeted cancer therapies occur, even in cohorts of molecular-defined homogenous cancers. Finally, by elucidating how the BIM deletion impairs the apoptotic response to TKIs, we now have an opportunity to overcome this resistance not only in CML but also other cancers.

Acknowledgements
We would like to acknowledge the many co-authors and collaborators who contributed their time and effort to the discovery of the BIM deletion polymorphism4), and the cancer patients in Japan, Malaysia, and Singapore, as well as their treating physicians, for providing samples for analysis.

The authors declare that they have no conflict of interest

References

1)Jänne PA, Gray N, Settleman J. Factors underlying sensitivity of cancers to small-molecule kinase inhibitors. Nat Rev Drug Discov. 2009; 8: 709-723.
2)Carella AM, Frassoni F, Melo J, et al. New insights in biology and current therapeutic options for patients with chronic myelogenous leukemia. Haematologica. 1997; 82: 478-495.
3)Schiller JH, Harrington D, Belani CP, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med. 2002; 346: 92-98.
4)Ng KP, Hillmer AM, Chuah CT, et al. A common BIM deletion polymorphism mediates intrinsic resistance and inferior responses to tyrosine kinase inhibitors in cancer. Nat Med. 2012; 18: 521-528.
5)Adams JM, Huang DC, Strasser A, et al. Subversion of the Bcl-2 life/death switch in cancer development and therapy. Cold Spring Harb Symp Quant Biol. 2005; 70: 469-477.
6)Cragg MS, Harris C, Strasser A, Scott CL. Unleashing the power of inhibitors of oncogenic kinases through BH3 mimetics. Nat Rev Cancer. 2009; 9: 321-326.
7)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. 2002; 100: 1014-1018.
8)Wheeler HE, Maitland ML, Dolan ME, Cox NJ, Ratain MJ. Cancer pharmacogenomics: strategies and challenges. Nat Rev Genet. 2012; 14: 23-34.
9)Mauro MJ. Defining and managing imatinib resistance. Hematology Am Soc Hematol Educ Program. 2006; 2006: 219-225.
10)Navin N, Kendall J, Troge J, et al. Tumour evolution inferred by single-cell sequencing. Nature. 2011; 472: 90-94.
11)Kantarjian H, Giles F, Wunderle L, et al. Nilotinib in imatinib-resistant CML and Philadelphia chromosome-positive ALL. N Engl J Med. 2006; 354: 2542-2551.
12)Kuribara R, Honda H, Matsui H, et al. Roles of Bim in apoptosis of normal and Bcr-Abl-expressing hematopoietic progenitors. Mol Cell Biol. 2004; 24: 6172-6183.
13)Kuroda J, Puthalakath H, Cragg MS, et al. Bim and Bad mediate imatinib-induced killing of Bcr/Abl leukemic cells, and resistance due to their loss is overcome by a BH3 mimetic. Proc Natl Acad Sci U S A. 2006; 103: 14907-14912.
14)Kubonishi I, Miyoshi I. Establishment of a Ph1 chromosome-positive cell line from chronic myelogenous leukemia in blast crisis. Int J Cell Cloning. 1983; 1: 105-117.
15)Egle A, Harris AW, Bouillet P, Cory S. Bim is a suppressor of Myc-induced mouse B cell leukemia. Proc Natl Acad Sci U S A. 2004; 101: 6164-6169.
16)Au WY, Caguioa PB, Chuah C, et al. Chronic myeloid leukemia in Asia. Int J Hematol. 2009; 89: 14-23.
17)Pao W, Miller VA. Epidermal growth factor receptor mutations, small-molecule kinase inhibitors, and non-small-cell lung cancer: current knowledge and future directions. J Clin Oncol. 2005; 23: 2556-2568.
18)The ENCODE Project Consortium, Birney E, Stamatoyannopoulos JA, et al. Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature. 2007; 447: 799-816.
19)Nakagawa T, Takeuchi S, Yamada T, et al. EGFR-TKI Resistance Due to BIM Polymorphism Can Be Circumvented in Combination with HDAC Inhibition. Cancer Res. 2013; 73: 2428-2434.
20)Delcuve GP, Khan DH, Davie JR. Roles of histone deacetylases in epigenetic regulation: emerging paradigms from studies with inhibitors. Clin Epigenetics. 2012; 4: 5.
21)Lee JK, Shin JY, Kim S, et al. Primary resistance to epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) in patients with non-small-cell lung cancer harboring TKI-sensitive EGFR mutations: an exploratory study. Ann Oncol. Prepublished on Apr 4, 2013, as doi: 10.1093/annonc/mdt127.
22)Katagiri S, Umezu T, Ohyashiki JH, Ohyashiki K. The BCL2L11(BIM)deletion polymorphism is a possible criterion for discontinuation of imatinib in chronic myeloid leukaemia patients. Br J Haematol. 2013; 160: 269-271.
23)Willis SN, Adams JM. Life in the balance: how BH3-only proteins induce apoptosis. Curr Opin Cell Biol. 2005; 17: 617-625.
24)Bouillet P, Huang DC, O'Reilly LA, et al. The role of the pro-apoptotic Bcl-2 family member bim in physiological cell death. Ann N Y Acad Sci. 2000; 926: 83-89.
25)Huntington ND, Puthalakath H, Gunn P, et al. Interleukin 15-mediated survival of natural killer cells is determined by interactions among Bim, Noxa and Mcl-1. Nat Immunol. 2007; 8: 856-863.
26)Sentman CL, Barber MA, Barber A, Zhang T. NK cell receptors as tools in cancer immunotherapy. Adv Cancer Res. 2006; 95: 249-292.
27)Helgason GV, Young GA, Holyoake TL. Targeting chronic myeloid leukemia stem cells. Curr Hematol Malig Rep. 2010; 5: 81-87.
28)Bhatia R, Holtz M, Niu N, et al. Persistence of malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete cytogenetic remission following imatinib mesylate treatment. Blood. 2003; 101: 4701-4707.
29)Goff DJ, Recart AC, Sadarangani A, et al. A Pan-BCL2 inhibitor renders bone-marrow-resident human leukemia stem cells sensitive to tyrosine kinase inhibition. Cell Stem Cell. 2013; 12: 316-328.
30)Gordon PM, Fisher DE. Role for the proapoptotic factor BIM in mediating imatinib-induced apoptosis in a c-KIT-dependent gastrointestinal stromal tumor cell line. J Biol Chem. 2010; 285: 14109-14114.
31)Will B, Siddiqi T, Jordà MA, et al. Apoptosis induced by JAK2 inhibition is mediated by Bim and enhanced by the BH3 mimetic ABT-737 in JAK2 mutant human erythroid cells. Blood. 2010; 115: 2901-2909.
32)Abrams MT, Robertson NM, Yoon K, Wickstrom E. Inhibition of glucocorticoid-induced apoptosis by targeting the major splice variants of BIM mRNA with small interfering RNA and short hairpin RNA. J Biol Chem. 2004; 279: 55809-55817.
33)Wertz IE, Kusam S, Lam C, et al. Sensitivity to antitubulin chemotherapeutics is regulated by MCL1 and FBW7. Nature. 2011; 471: 110-114.
34)High LM, Szymanska B, Wilczynska-Kalak U, et al. The Bcl-2 homology domain 3 mimetic ABT-737 targets the apoptotic machinery in acute lymphoblastic leukemia resulting in synergistic in vitro and in vivo interactions with established drugs. Mol Pharmacol. 2010; 77: 483-494.
35)Nakagawa T, Takeuchi S, Yamada T, et al. EGFR-TKI Resistance Due to BIM Polymorphism Can Be Circumvented in Combination with HDAC Inhibition. Cancer Res. 2013; 73: 2428-2434.

1Cancer & Stem Cell Biology Signature Research Programme, Duke-NUS Graduate Medical School, Singapore
2Department of Haematology, Singapore General Hospital, Singapore
3Department of Medical Oncology, National Cancer Centre, Singapore
4Division of Medical Oncology, Department of Medicine, Duke University Medical Center Durham, NC, USA

前へ戻る