Background

TP53 mutations inactivating p53 protein, often associated with loss of the remaining TP53 allele through 17p deletion, are major prognostic factors in many hematological malignancies, including CLL, myeloma, AML and MDS. In AML and MDS, they are usually associated with complex karyotype (including del 17p) and very poor prognosis (Blood 1991, 78(7):1652-7 , Bejar, NEJM 2011), including after allogeneic SCT (Middeke JM, Blood 2014) but they are also seen in lower risk MDS with isolated del 5q, where they confer resistance to Lenalidomide (Jadersten, JCO 2011). The advent of Next Generation Sequencing (NGS) techniques has improved the detection of such mutations, by allowing the identification of small mutated clones. Other detection methods may prove interesting, especially functional methods like FASAY ( Functional Assay of Separated Allele in Yeast) , an easy and sensitive method that detects TP53 mutations by assessing the p53 function as transcription factor (Flaman et al, PNAS 1995). We compared the detection of TP53 mutations in MDS and AML by FASAY and NGS approaches.

Methods

The 84 patients analyzed included 10 AML, 10 higher risk MDS, and 64 lower risk MDS with del 5q. RNA and DNA were extracted from marrow mononuclear cells. TP53 mutations were detected on RNA by FASAY where, after amplification of the TP53 mRNA, the PCR product is co transfected with an open gap repair plasmid leading, by homologous recombination, to p53 protein expression in the yeast. The yeast strain used is dependent on p53 functionality for growth and color and detection of more than 10% of small red yeast colonies defines a non-functional FASAY result. All non-functional FASAY were confirmed by the split versions of the test and TP53 defects were characterized by Sanger sequencing. The detection limit is around 10% in our hands (Manie E, Cancer Res 2009). In parallel, TP53 mutations were detected on DNA by NGS using the IRON II plate design and pyrosequencing on a GS Junior System (Roche). (Kohlmann, Leukemia 2011).FASAY (+Sanger sequencing) and NGS were performed in two different labs.

Results

By FASAY, 47 patients (56%) had a functional p53 and 37 cases (44%) a non-functional p53 and a mutation was confirmed by Sanger in all non functional cases.

By NGS analysis, no TP53 mutation was found in 47 cases (56%) and a mutation was detected in 37 cases (44%).

In the 37 mutated cases by NGS, the median proportion of mutated allele was 35% (range 3 to 99%), including a median of 72%, 35%, 25 % in AML, higher risk MDS and lower risk MDS with del 5q, respectively. The mutated clone size was lower than 10% in only 2 patients who both had lower risk MDS with del 5q (3 and 6%, respectively).

A perfect correlation between FASAY and NGS was found in 80 (95.5%) cases. The 4 discordant cases included a mutation detected only by FASAY in 2 cases, and only by NGS in 2 cases. Undetected mutations by NGS were insertions of intronic sequences (intron 9) not explored by the technique used. These insertions resulted in non-functional protein well detected by FASAY which analyses the global cDNA sequence including splicing defects. Undetected mutations by FASAY were mutations in which the percentage of mutated alleles was less than 10% (3% and 6 % respectively).

Finally, while the cost of NGS analysis for TP53 mutation is around 200 euros when performed alone (and around 2000 euros when combined to analysis of the 30 main other genes involved in MDS and AML), the cost of the FASAY technique is around 20 euros (prices including reagents only).

Conclusion

The FASAY technique is a cheap method, that in spite of a sensitivity of only 10%, was able to detect 98% of TP53 mutations detected by NGS. In fact those mutations appear to involve generally relatively large clones in MDS and AML. FASAY could also detect 2 atypical intronic mutations overlooked by NGS. Demonstrating in such difficult cases that the resulting p53 protein is non functional and therefore probably has pathophysiological significance, is an advantage of FASAY .The combination of the 2 methods, and especially the combination of DNA and RNA analysis, may be useful in such cases.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution