Page 24 - Read Online
P. 24

Page 24                                                Kaehler et al. Cancer Drug Resist 2019;2:18-30 I http://dx.doi.org/10.20517/cdr.2019.05

               becomes more likely. Furthermore, bioavailability of drugs is impaired influencing adsorption, distribution
               and elimination. The most prominent transporters regarding chemoresistance are P-glycoprotein (P-gp,
               ABCB1), breast cancer resistance protein (BCRP, ABCG2) and members of the ABCC-family, especially
               ABCC1/MRP1 (multidrug resistance protein 1), ABCC2/MRP2 (multidrug resistance protein 2) and ABCC3.
               Drug resistance via drug transporters is not only influenced by deregulation of gene expression, but also by
                            [66]
               genetic variants . While a vast number of studies showed impact of genetic variants on drug transporter
               expression or activities, recent data dealt with variation of 3’-UTR length in drug resistant cancer cells.

               P-gp in drug resistance
               A huge number of classical and targeted chemotherapeutics are substrates for P-gp and many studies
               demonstrated an overexpression of P-gp upon exposure to such drugs [Table 1]. Consequently, a potential
               impact of genetic variants on ABCB1 expression and subsequently in response to treatment could be found
               with more than 4,453 SNPs identified in this gene. Most clinical studies focused on 3435C>T, 2677G>T/A and
               1236C>T that were suggested to impair P-gp expression or activity.


               The cytosine deamination at position 3435 (rs1045642) in exon 26 does not result in an amino acid
               substitution. Nevertheless, a number of studies suggested a lower P-gp expression and reduced P-gp function
               in 3435T carriers [67,68] . In acute lymphocytic leukemia, carriers of the wildtype revealed higher toxicity after
               high-dose methotrexate treatment, while the risk of relapse was reduced in carriers of at least one variant
                    [69]
               allele . Nonetheless, a direct association to cancer therapy failure could not be seen. The impact of the
               silent SNP 1236C>T (rs1128503) in exon 12 instead is not entirely understood, as contradictory studies
               showed opposite results regarding response to imatinib in chronic myeloid leukemia [70,71] . In the same
               studies, amino acid substitution of alanine to serine/threonine due to 2677G>T/A (A893T, rs2032582) could
               be associated to imatinib response, with contradictory effects as carriers of the CC variant revealed altered
               susceptibility towards imatinib therapy compared to wildtype carriers. Regarding missense mutations in
               ABCB1, an experimental approach using Saccharomyces-based assay revealed several non-synonymous
               SNPs in the context of chemoresistance, e.g., 266T>C (M89T), 1985T>G (L662R), 2005C>T (R669C) 3322T>C
               (W1108R) and 3421T>A (S1141T) [72,73] . Nevertheless, clinical relevance of these SNPs has to be confirmed.
               Overall, the issue on ABCB1 variants and their association to drug response remains controversial. Due to
               its high phenotypic variability however, it is more than questionable to use ABCB1 variants to as predictive
               biomarker in cancer therapy. In addition to hereditary variants, the regulation of ABC transporters by
               nuclear receptors, cytokines and also microRNAs is of major importance. In this regard, differential
               expression of 3’-UTR lengths of the ABCB1 mRNA might additionally contribute to P-gp variability (see
               epigenetics section).


               ABCG2/BCRP and chemoresistance
               Besides P-gp, BCRP was identified as a contributing factor to multidrug resistance [Table 1]. Being highly
                                                                                              [75]
                                                             [74]
               expressed in hematopoetic precursor and stem cells , it is regarded as a stem cell factor . Moreover,
               its high expression in tumor cells pointed to a potential role in chemoresistance and therapy failure [76,77] .
               Main research has been performed on non-synonymous 34G>A and 421C>A. The SNP 34G>A (rs2231137)
               results in exchange of valine to methionine (V12M) at the N-terminus potentially leading to reduced
               protein expression. This SNP could be associated to outcome of tyrosine kinase inhibitor therapy, especially
                            [78]
               using imatinib . Some clinical evidence point to a better response to imatinib in CML patients carrying
                                    [78]
               the homozygous variant . A similar tendency was observed in metastatic renal cell cancer treated with
                       [79]
               sunitinib . On the contrary, an association of therapy using erlotinib in B cell lymphoma patients could
                            [80]
               not be observed . The 421C>A substitution (rs2231142) leads to an amino acid exchange from glutamine to
               lysine (Q141K) that affects the ATP binding domain. In a number of studies, bioavailability of drugs, e.g., the
                                                                                  [82]
               topoisomerase inhibitor topotecan  or the tyrosine kinase inhibitor imatinib , was impaired in carriers
                                            [81]
               of the 421C>A genotype. Regarding tyrosine kinase inhibitors, many compounds are substrates of BCRP,
               however transport is dose-dependent . Interestingly, the 421A haplotype could be associated to modulate
                                               [83]
   19   20   21   22   23   24   25   26   27   28   29