Page 113 - Read Online

Abaji et al. Cancer Drug Resist 2019;2:242-55 I http://dx.doi.org/10.20517/cdr.2018.24 Page 251

rs11554772, and a higher risk of ALL relapse in patients who received E. coli ASNase. This gene codes for
a transcriptional factor involved in ASNS gene regulation. Importantly, the result was validated in the
replication group and was corroborated with data on the association of the same polymorphism with higher
promoter activity. Another finding was the association of a 14-bp tandem-repeat polymorphism, rs3832526,
located in the first intron of ASNS gene and EFS which showed that homozygous carriers of the double
[18]
repeat (2R) had a significantly lower EFS, but the association lacked significance in the validation cohort .
They also reported the association of polymorphisms in the ASS1 gene and EFS, albeit these associations did
[18]
not sustain correction for multiple testing and thus were not further investigated in the replication cohort .
Interestingly, the repeat polymorphism in ASNS gene was later linked to early response to ALL treatment
following the administration of a single ASNase dose in a study of 264 Polish children with ALL. However,
the association was in the opposite direction as carriers of the (3R) allele with a poor response at day 15 had
an increased risk of events, hence the data suggest an interaction between this polymorphism and early
[54]
response to treatment that could result in variability of EFS rates .
MicroRNA
One area that is currently under-investigated in relation to the effect of ASNase is that of microRNAs
(miRNA), with only few studies reporting associations between differences in miRNAs expression levels and
response in childhood ALL [55-57] . While reports suggest that the expression of over 60% of protein coding
[58]
genes is subject to regulation via miRNAs . Of note, many groups linked the expression levels of specific
miRNAs to clinical outcome of ALL patients. In fact, studies suggest that miRNA expression profiles can
differ significantly between ALL genetic-subtypes and that drug-resistant cases are associated with unique
miRNA signature. For example, one study showed that miR-454 was expressed at nearly two-fold lower
[55]
levels in ASNase-resistant pediatric ALL patients when compared to ASNase-sensitive ones . Another
study linked miR-210 to ASNase-sensitivity as demonstrated by the expression levels dependent change in
the minimum inhibitory concentration (IC ), the concentration needed to block the proliferation of half of
50
the initial cell population .
[56]

CONCLUSION
While it is becoming increasingly recognized that both tumor and germline genomics can influence
response to treatment, the latter is less commonly used to guide treatment in oncology settings. It should be
emphasized that the possibility of detecting a random signal in association studies is relatively high, which
could explain the conflicting data and inconclusive results among studies that targeted the same genetic-
phenotypic associations. Moreover, differences in trial settings, treatment protocols, nature of supportive
care, the degree of scrutiny with which an outcomes is measured and variations in disease characteristics,
among others, can influence the role of the variant in question [10,12] . One example is the leukemic cells that
carry the subtype of ALL featuring a TEL/AML1 fusion gene which were demonstrated to be more sensitive
[59]
to the effect of ASNase compared to other subtypes . Thus, the implication of a gene or its polymorphisms
in the outcome should only be taken into consideration for clinical implementation if the association was
confirmed by independent studies and further supported by functional analysis.

The translatability of pharmacogenetics findings into the clinical realm of personalized medicine remains
a challenge given the complex interplay between the host and malignancy genomes. One example is
the CoALL 06-97 study which incorporated a combined drug resistance profile into their risk group
stratification process of 224 patients. While this profile, which was based on in vitro cellular resistance
to prednisolone, VCR and ASNase, was previously shown to be linked to treatment response and was
confirmed in several studies, the authors reported no significant difference between results of that study and
[60]
those of historical control group stratified according to conventional risk factors . A lot of work needs to
be done in the context of implementation of pharmacogenetics. In a study that analyzed pharmacogenomics