Serum Free Light Chains Assay
for Response Assessment in
Myeloma
Shaji kumar, MD
Mayo Clinic
Rochester
Methods
· 34 patients were enrolled in the trial; 27 patients who
had serial FLC assessments were studied.
· FLC estimation was carried out using the serum FLC
assay (FreeliteH, The Binding Site Limited, UK)
performed on a Dade-Behring Nephelometer.
· monoclonal FLC: / FLC ratio <0.26
· monoclonal FLC: / FLC ratio >1.65
Response Definitions
· Partial response (PR) required an abnormal baseline
FLC ratio and any one of the two following criteria:
· 50% decrease in the level of the involved FLC plus a 50%
decrease (or normalization) in the ratio of
involved/uninvolved FLC
or
· 50% decrease in the difference between involved and
uninvolved FLC levels.
· Complete response (CR) required normalization of
FLC ratio and negative serum and urine
immunofixation.
Patients
· Response at 4 months or earlier by the Blade criteria
was compared to FLC response criteria from the same
evaluation.
· Three patients had normal FLC levels and ratio at base
line and were not included in the analysis.
Dynamics of light chain response (kappa = involved light chain)
350
300
250
seline
Kappa
200
ba
Lambda
of
Ratio
150
tage
Difference
100
ercenP 50
0
01m
2m
3m
4m
Time from start of therapy
Dynamics of light chain re sponse (lambda = inv olv e d light chain)
300
250
ein 200
asel
kappa
fb
lambda
150
eo
ratio
tag 100
difference
ercenP 50
0
0
1m2m
3m4m
Time from start of therapy
PR by M protein
PR by FLC
PR-by FLC ratio
PR by FLC difference
7
6
5
erb 4mu
en 3
yclC
2
1
0
01
2
3
4567
89
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Patient number
Results
· 23 of the remaining 24 achieved a PR or better by
Blade criteria.
· A 50% decrease in the level of the involved FLC + a
50% decrease (or normalization) in the ratio of
involved/uninvolved FLC correctly classified 20 of the
22 responding pts (sensitivity 91%)
· 2 patients could not be evaluated since baseline FLC ratio could not
be calculated.
· On the other hand, a 50% decrease in the
mathematical difference between involved and
uninvolved FLC levels correctly classified all 24
responding patients (sensitivity 100%).
Results
· The one non-responding patient by Blade criteria was
correctly classified by both FLC criteria.
· All patients were correctly classified by both criteria
when only those with a baseline involved FLC level of
at least 10mg/dL (100mg/L) were considered (15
patients).
· In patients the PR by FLC criteria was attained at 1-2
cycles following PR by conventional criteria
Conclusions
· A 50% decrease in the difference between involved
and uninvolved FLC levels will suffice as FLC criteria
for PR, eliminating the need for the alternative criteria
based on the involved FLC level and the ratio.
· This FLC response criteria should be used in patients
not having measurable levels of serum and /or urine M
· This study is limited by the lack of adequate non-
responders to calculate specificity and lack patients
who progressed to validate progression criteria and
needs further validation.