Oral leukoplakia (OL) and oral submucosal fibrosis (OSMF) are precancerous conditions with common etiologies but with different risks for oral cancer (OC) progression. In rare cases, both conditions occur in the same patient and provide an opportunity for understanding the common and distinctive variants upon exposure of genetically identical normal cells to the same carcinogen(s). We performed exome sequencing of a patient with OL (hyperplasia, but no dysplasia) and OSMF (grade II) in the opposite cheeks using blood DNA as the reference genome. The overall somatic variant burden was higher in OSMF than OL, but opposite in the case of copy number alterations. OL-specific variants were enriched in genes associated with DNA repair, cell division/cell cycle checkpoint pathways, whereas in OSMF, extracellular matrix-receptor interaction was mainly affected. The proportions of variants in cancer driver genes and cancer driver mutations were similar in both cases indicating no difference in the potential risk associated with the two conditions at the stages sampled. Future studies on rare cases similar to the one described in this report will help in understanding the molecular basis of differences associated with OL and OSMF and shared processes accompanying OC progression.

BACKGROUND: Osteosarcoma is a highly malignant tumour associated with numerous and complex genetic alterations like copy number alterations. Recent whole genome studies revealed distinct mutations in several candidate oncogenes. While clinical parameters stratify osteosarcoma patients in risk groups, genetic profiles have not yet been used to tailor tumour treatment. However, specific copy number alterations seem to have a prognostic impact in osteosarcoma treatment. Somatic TP53 gene mutation frequently occurs in sporadic osteosarcoma. When arising germline, TP53 mutation leads to Li-Fraumeni syndrome and may result in early life osteosarcoma. The effect of Li-Fraumeni syndrome on the genetic profile of osteosarcoma and the consideration of the syndrome during cancer treatment are topics of current research.
METHODS: We report a 25-year-old female with pelvic osteosarcoma refusing continuation of therapy. She interrupted neo-adjuvant chemotherapy according to EURAMOS-1/COSS recommendations and declined local or further adjuvant therapy. Surprisingly, she remained in sustained remission for the osteosarcoma but eventually died from newly diagnosed breast cancer. After establishment of breast cancer, we detected TP53 germline mutation and investigated the osteosarcoma material with array-CGH.
CONCLUSIONS: Genetic examination of the tumour evidenced several copy number alterations with striking differences to previously reported data. We discuss possible influences of the genetic profile on the unusual clinical course and the significance of Li-Fraumeni syndrome for the genetic profile. Specific loss of (proto-) oncogenes might have contributed to the unusual case. Further large-scale genetics of Li-Fraumeni patients combined with detailed clinical data will help to identify specific genetic risk profiles and improve treatment.

Refining risk stratification to avoid very early relapses (VER) in Mexican patients with B-lineage acute lymphoblastic leukemia (B-ALL) could lead to better survival rates in our population.
The purpose of this study was to investigate the association between the United Kingdom ALL (UKALL)-CNA classifier and VER risk in Mexican patients with childhood B-ALL.
A nested case-control study of 25 cases with VER and 38 frequency-matched controls without relapse was conducted within the MIGICCL study cohort. They were grouped into the categories of the UKALL-CNA risk classifier (good [reference], intermediate and poor), according to the results obtained by multiplex ligation dependent probe amplification. Overall and disease-free survival (DFS) were estimated using the Kaplan-Meier method. Univariate and multivariate Cox proportional hazards analyses were conducted.
The CDKN2A/B genes were most frequently deleted in the group with relapse. According to UKALL-CNA classifier, 33 (52.4%) patients were classified as good, 21 (33.3%) intermediate and 9 (14.3%) poor-risk B-ALL. The intermediate and poor risk groups were associated with an increased risk of VER (HR = 4.94, 95% CI = 1.87-13.07 and HR = 7.42, 95% CI = 2.37-23.26, respectively) in comparison to the good-risk patients. After adjusting by NCI risk classification and chemotherapy scheme in a multivariate model, the risks remained significant.
Our data support the clinical utility of profiling CNAs to potentially refine current risk stratification strategies of patients with B-ALL.