Microarray CGH Analysis of Circulating Tumoral Plasma DNA in NF1 Patients With MPNSTs

Introduction Neurofibromatosis type 1 (NF1) is an autosomal dominant disorder occuring in 1 out of 3500 living newborns. The disease is caused by heterozygous mutations of the NF1 gene, located on chromosome 17q11.2. The NF1 gene encodes the tumor suppressor neurofibromin, a negative regulator of the RAS oncogene. Clinically, NF1 patients have café-au-lait maculae, freckling, Lisch nodules and neurofibromas. Cognitive problems, bone lesions and optic pathway gliomas are also common in these patients.

Neurofibromas are benign neoplasms of the peripheral nerve sheath, which can appear anywhere in the body of NF1 patients. Neurofibromas are composed of different cell types like Schwann cells, fibroblasts, mast cells and perineurial cells. If in Schwann cells of NF1 patients the wild-type NF1 allele is inactivated, a neurofibroma will be formed. There are 3 types of neurofibromas: cutaneous, subcutaneous and plexiform. Cutaneous neurofibromas appear during adolescence as isolated nodules in or under the skin, respectively. Plexiform neurofibromas are congenital and can spread along a large segment of a peripheral nerve. Subcutaneous neurofibromas are discrete nodules located on peripheral nerves and the timing of their origin is not known, but they are probably also prenatal in origin.

Every NF1 patient has a life time risk of 8 to 13% to develop a malignant peripheral nerve sheath tumor (MPNST). This risk is even twice as high for a subpopulation of NF1 patients, namely patients with an NF1 microdeletion. MPNSTs are difficult to diagnose in early phase because of the large number of tumors and diverse locations often seen in these patients. These tumors infiltrate easily in surrounding tissue and frequently give rise to metastases. At this moment, the only available treatment is surgical removal of these MPNSTs. Complete resection is often difficult due to the size of the tumors and the location on important nerves and the fact that the tumors are frequently already metastasized at diagnosis. The five-year survival of patients with MPNSTs is less than 25%. For MPNSTs, like for other cancers, the extent and the spread of the disease at time of diagnosis is an important factor in determining treatment outcome. In this regard, tumor markers have been developed to screen for tumors, but only a subset of cancers secrete specific proteins that can be used as a marker. An increasing interest grew in new markers like circulating nucleic acids. The detection of cell-free circulating nucleic acids in plasma, serum and other body fluids of healthy and diseased individuals opened up the possibility to diagnose and monitor the disease, which might also be applicable to NF1 patients with MPNSTs. The presence of extracellular circulating nucleic acids in the blood of healthy and sick persons was first described by Mandel and Métais in 1948. In 1977 the group of Leon was the first to quantify cell-free DNA in serum. They found that cancer patients tended to have elevated levels of circulating DNA compared to healthy control individuals, and that the amount of DNA was even bigger in patients with metastases compared to localized disease, but nothing was known about the origin of the DNA. Significant progress in plasma/serum DNA research was made, when the presence of tumor-associated microsatellite alterations was reported in the plasma and serum of cancer patients. According to these findings, it has been suggested that the circulating DNA could have potentials for the diagnosis and prognosis of malignancies. Still, the mechanism of DNA release is not fully clarified, and different hypotheses have been suggested. A first hypothesis that the circulating DNA originates from DNA leakage due to necrosis or apoptosis of tumor cells, became controversial when it was reported that the DNA level decreased after radiotherapy. Another hypothesis about micrometastases was also rejected because the number of cells in plasma didn't match the amount of circulating DNA. A third hypothesis is the spontaneous and active release of tumoral DNA in the blood by proliferating cells. However there is no convincing explanation for the latter mechanism.

Research protocol A first part of the study will be the optimization of the technique of DNA extraction from plasma. Different DNA extraction methods described in the literature will be tested on samples collected in cooperation with Prof. Dr. Patrick Schöffski (UZLeuven). When blood is taken from cancer patients for routine testing, an additional 5 mL (in EDTA tube) will be available for our experiments. Blood samples will be collected within a few hours after sampling and processed within 24 hours in our lab (lab for neurofibromatosis research). The blood samples will be centrifuged at 1600 rpm for 10 minutes. The supernatant plasma will be collected, without disturbing the buffy coat, and re-centrifuged at 13000 rpm for 10 minutes to remove particles. The supernatant plasma will be aliquoted into 0.5 mL volumes. The samples will be stored at -80°C until further use. Per patient, different aliquots will be extracted using different methods, like phenol-chloroform extraction and commercial kits, according to the literature. Genomic reference DNA will be extracted from the buffy coat using phenol-chloroform extraction. Depending on the type of DNA extraction, we expect differences in quantity and quality of the DNA. The method with the highest yield and best DNA quality will be chosen to extract DNA from plasma of NF1 patients with MPNSTs.

Second part of this work will be the optimization of the microarray experiment using plasma DNA. Plasma samples from NF1 patients with MPNSTs will be collected in collaboration with the group of Victor Mautner from the Eppendorf University in Germany. Direct processing of blood samples will be done in Germany and frozen buffy coats and plasma aliquots will be sent to Leuven in an encoded way. Together with the buffy coats and plasma, frozen tissue of the patients MPNST will be sent as well. DNA from the tumor will be extracted using phenol-chloroform extraction. Microarray CGH will be performed on both the tumor and the plasma DNA with buffy coat DNA as reference to exclude copy number variable regions from the list of copy number changes. The array profiles of the plasma DNA will be checked for the tumor-related copy number changes. In that way, the array protocol can be further optimized in order to detect tumor specific copy number abnormalities in the cell-free DNA from plasma.

This study is performed by Eline Beert, PhD student in the lab for neurofibromatosis research, and Radost Ratcheva, a master thesis student in the lab of neurofibromatosis research, as part of achieving the degree of 'Doctor in Biomedical Sciences' and 'Master in Biomedical Sciences', respectively, with Prof. Dr. Eric Legius as promotor and main researcher. Only the Katholieke Universiteit Leuven and the UZ Leuven, but no commercial partners (like companies) will participate in this study. The results obtained from the described experiments have to be considered as pure scientific information and none of it will be reported to the patients.

The blood samples of the UZLeuven cancer patients will only be used to optimize the DNA extraction procedure, the extracted DNA will be quantified but it will not be further analyzed and will be destroyed afterwards.