Once believed to be a homogeneous disease entity, multiple myeloma now represents epitome of heterogeneous genomic evolution and mutational profiles. Various groups have reported different risk stratification methodologies based on cytogenetic data, especially the fluorescent in situ hybridization (FISH) results. Reflecting the racial disparities, Oh et al. have previous described prognostic model specific to Koreans using a nationwide registry. However their main focus has been on conventional cytogenetics. We sought to analyze the predictive and prognostic value of FISH on top of conventional cytogenetics from a large cohort of Korean multiple myeloma patients.From a single tertiary academic center, 1006 newly diagnosed MM patients were retrospectively identified during the period of January 2005 to June 2015. Adult patients, defined as 18 years old or older were included, while cases with smoldering myeloma, monoclonal gammopathy, solitary plasmacytoma and plasmoblastic leukemia were excluded in the first place. After excluding additional 441 patients for insufficient data, a total of 565 patients with complete set of molecular information were evaluated. Progression free survival (PFS) was defined as the duration from the start of first line treatment to disease progression or death, and data available up to June 2016 were used. On a different note, Korea has an unique public medical insurance system that is mandatory and covers approximately 98% of the overall Korean population Although the insurance covers comprehensive aspects of healthcare, the range of coverage is restricted. In this study, autologous stem cell transplantation (autoSCT) eligible patients were defined as those under the age of 65 years according to such insurance coverage restrictions. For the same reason, first line treatment were diverse according to the patient's financial standings.The median age was 63 years (18-92 years), and there were 309 males (54.7%). As for MM subtype, IgG type was 49.0%, IgA type 18.1%, and light chain type 28.8%. The proportion of patients at ISS stage III was 32.2%, and 45.8% had poor performance status (ECOG 2 or more). Overall, the median line of treatment was 2 (range 1-16) and 42.7% of the patients underwent autoSCT. The patients who actually underwent autoSCT had significantly better PFS compared to those who did not (not reached vs 18 months, P <0.001). The median PFS for each first line chemotherapy regimen are listed in Table. For autoSCT ineligible patients, novel agents such as lenalidomide and carfilzomib were more frequently used and this translated into surprisingly long median survival of 51 months.FISH results showed p53 deletion in 9.1% (27/296), 13q14 deletion in 36.0% (174/484), IgH rearrangement in 47.6% (231/485), and 1q21 amplification in 40.0% (193/483) of cases. Those with p53 had the worst prognosis with median PFS at 13 months, and its presence significantly altered overall PFS (P=0.002). This phenomenon was observed both in autoSCT eligible group (PFS for p53 deletion present 20 months vs not present 77 months, P<0.001), and in ineligible group (PFS for p53 deletion present 7 months vs not present 24 months, P<0.001). The presence of p53 was associated with reduced bortezomib sensitivity (P=0.003), but not thalidomide. On the other hand, the presence of 13q14 deletion was not significant for PFS in autoSCT eligible group (P=0.726), but was significant in ineligible group (P=0.023). The presence of 13q14 was associated with reduced sensitivity to MP/VAD regimen (P=0.005). The presence of 1q21 did not have survival implications or treatment predictive values. Likewise, IgH rearragements did not affect PFS and were not useful for individualizing therapy.MM patients in Korea presents a unique set of population associated with different survival benefits per therapy compared with other ethnicities. Since FISH data can be readily used to predict the treatment outcomes of particular agents, a panel markers specific to Korean race should ensue to establish risk-adaptive therapeutic approaches for equitable distribution of limited medical resources. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.