Introduction: Identification of patients
with high-risk (HR) multiple myeloma (MM) is important to optimise their
treatment. In 2015, revised International Staging System (R-ISS)
guidelines were published (Palumbo, 2015) where HR cytogenetics (CG) and
elevated serum lactase dehydrogenase (LDH) were added to the
traditional ISS staging criteria. Thus, R-ISS stage III includes one of
the HR CG abnormalities or elevated LDH, ISS stage I patients have no HR
CG and normal LDH; the rest of patients belong to R-ISS stage II. There
are limited data on the prevalence of R-ISS groups in comparison to the
old ISS grouping and impact on clinical outcomes in the real-life
setting. The aim of the present analysis was to use structured
longitudinal electronic health records (EHR) provided by the Finnish
Auria Biobank to compare the prevalence and survival outcome between
patients in ISS and R-ISS groups in a real-life patient cohort of 100
patients treated at Turku University Hospital. Auria Biobank covers
roughly 15% of the population of Finland and collects samples with the
associated data from all diseases treated at Turku University Hospital
based on the Finnish Biobank Act.

Methods:
Auria Biobank database was analysed retrospectively for all MM-patients
diagnosed between 2008-2013 (incident cohort) and whose fluorescence in
situ hybridization analysis was performed at the time of diagnosis.
Data for age, gender, LDH, creatinine, ISS, R-ISS, CG, time to next
treatment and overall survival were collected from Auria Biobank.
Classification into ISS groups was done based on data at the time of
diagnosis and OS. Classification into R-ISS staging was done according
to IMWG (Palumbo, 2015). For HR CG at least one of the following CG
abnormalities was required: del(17), t(4:14), or t(14:16). Estimated
glomerular filtration rate was calculated by using the CKD-EPI formula.
Drug treatments were classified as conventional (e.g. melphalan +
prednisolone) or novel (proteasome inhibitors, IMIDs). Descriptive
methods and Kaplan-Meier survival analysis were used for comparison of
the groups.

Results: The median age of
the 100 patients was 64 yrs (range: 37 - 80), and 43% were female. At
the time of diagnosis, 17% of patients had high risk CG status, 32% had
at least moderate kidney failure (estimated glomerular filtration rate
<60ml/min) and 26 % had elevated LDH. 41% patients received
autologous stem cell transplant and 64% and 14% were treated with novel
and conventional treatments in first line, respectively. 26%, 48% and
26% were classified to ISS stage I, II and III groups respectively, and
21%, 63% and 16% to R-ISS stage I, II and III groups, respectively.
Criteria to include patients into R-ISS III were HR CG in 62 %, elevated
LDH in 69 % and 31 % fulfilled both criteria. Neither ISS- nor R-ISS
staging had any influence on first line treatment decisions between
novel and conventional treatments.

In all patients 2-year
OS (from diagnosis) was 82% (median OS not yet reached). The 2-year OS
in ISS I, II and III groups was 90%, 88%, and 60% respectively, and in
R-ISS I, II and III groups 87%, 85% and 42%, respectively. R-ISS had a
statistically significant effect on survival time (log rank P=0.032),
with R-ISS III patients having a 3.8-fold risk of death compared to
R-ISS I (Fig 1). R-ISS III patients had also shorter (ns) treatment free
survival than R-ISS I patients (HR 2.1, log rank P=0.479) (Fig 2). No
statistically significant difference was observed between the survival
curves stratified by ISS staging groups.

Conclusion:
The new R-ISS staging system, using additional information including
CG-profile and serum LDH, separated in our real-life setting more
profoundly patients with poor prognosis than the old ISS staging.
Structured EHRs can successfully be used to derive useful clinical and
prognostic data from real-life MM patients.