Reduced Expression Level of SHP1 Gives An Additive Survival Advantage to the Ph+ Cells of Chronic Myeloid Leukemia (CML) Patients and Provides a Novel Pretreatment Predictor of Major Molecular Response Achievement in CML Patients.

Abstract 2212

Poster Board II-189

Treatment of Chronic Myeloid Leukemia (CML) has shown an outstanding progress while new understanding of the disease has increased significantly. Nevertheless in the era of targeted therapy, Sokal index remains a dominant prognostic determinant of newly diagnosed CML patients. Our study has aimed to identify novel prognostic indicators to improve both an initial assessment and subsequent monitoring of CML patients. Initially, we have found that the protein-tyrosine phosphatase SHP1, that has a tumour suppressor activity, may play an important role in the resistance to imatinib treatment. We applied gene profiling and proteomic bidimensional electrophoresis to compare the differential pattern of gene and protein expression between KCL22s (imatinib-sensitive) and KCL22r (imatinib-resistant) cell lines. We found SHP1 to be one of the most differentially expressed genes. By ESI-TRAP MS technique, we found that one of the main interactors of SHP1 is SHP2, a protein phosphatase well known as positive regulator of oncogenic pathways, including the Ras/MAPK pathway. Gain-of-function mutations in SHP2 gene, have been described in various haematopoietic neoplasias and myeloproliferative disorders including Juvenile Chronic Myelomonocytic Leukemia. This protein is regulated throw phosphorylation on 542- and 580-Tyr, and unlike SHP1, acts as a positive regulator of the same oncogenic pathways. We found that KCL22r cell line, that has low SHP1 levels, showed complete phosphorylation of both SHP2 tyrosine residues, while these residues are not phosphorylated in the KCL22s line, which could explain an important mechanism for imatinib sensitivity. Consistently with this hypothesis, knock-down of SHP2 phosphatase in KCL22r by a specific shRNA resulted in 60% inhibition of KCL22r proliferation. Furthermore, the KCL22r^SHP2- cells showed significant reduction in STAT3 (60%) and ERK1/2 (70%) phosphorylation. Our initial results from CML patients (Esposito et al , ASH 2008 Abs 1106) have suggested a differential expression of SHP1 in patients with different response to imatinib treatment. To further explore the role of SHP1 as a determinant of imatinib sensitivity we evaluated the expression of SHP1 in 93 newly-diagnosed CML patients enrolled into the TOPS trial investigating 400mg versus 800mg imatinib (Cortes et al, EHA 2008). The results of this study indicate that the mRNA levels of SHP1, as assessed by QPCR in peripheral blood of patients at the time of enrolment, are significantly different between patients who do or don't achieve Major Molecular Response (MMR) by 12 months (7.9±4.0 vs. 5.9±3.4; p=0.01). Logistic regression was used to estimate regression coefficients and corresponding odds ratio using MMR by 12 months as outcome variable in our model. Since the 25^th and 75^th percentiles of SHP1 were 4.3 and 8.4, respectively (resulting in an interquartile range of 4.1), statistical analysis shown that a value of 4.1 or more in SHP1 is associated with almost 2-fold odds of achieving MMR by 12 months (OR=1.92; 95% CI=1.12, 3.29; p=0.018). Moreover, in a contingency table, chi-square analysis has been shown a high risk of not achieving MMR by 12 month in those patients with either low SHP1 expression and high Sokal score, when compared with patients with high-intermediate SHP1 expression and low-intermediate Sokal score (p=0.0068).

In conclusion, these results suggest that, measuring expression levels of SHP1 could be of value in assessing newly diagnosed CP-CML patients and estimating treatment response, which could help optimizing Gleevec treatment, or recommending patients to more potent TKIs.

Supported by Novartis Oncology, Clinical Development, TOPS Clinical Correlative Studies Network