HAL Id: hal-03261886
https://hal.inria.fr/hal-03261886
Submitted on 16 Jun 2021
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Quantitative modeling of metastasis: cancer at the
organism scale
Sébastien Benzekry
To cite this version:
Sébastien Benzekry. Quantitative modeling of metastasis: cancer at the organism scale. Doctoral.
United States. 2021. �hal-03261886�
Quantitative modeling of metastasis: cancer at the organism scale
S. Benzekry
Head of the Inria-Inserm team COMPO
Marseille, France
MATHEMATICAL MODELING
J. Ciccolini
R. Fanciullino
A. Rodallec
Mechanistic
modeling
Pharmacometrics
Statistics
Data science
D. Barbolosi
S. Benzekry
D. Barbolosi, S. Benzekry
F. Gattacceca
S. Benzekry
PHARMACOLOGY
Experimental
J. Ciccolini
R. Fanciullino
B. Lacarelle
Clinical
MEDICINE
Pr L. Greillier Dr X. Muracciole
Pr S. Salas
Clinical problem
8
<
:
ˆ
tfl
(t, v) + ˆ
v(
g(t, v)
fl
(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
fl
(0, v) = 0
g
(v)
= (
–
≠
—
ln(v)) v
8
<
:
ˆ
tfl
(t, v) + ˆ
v((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
fl
(0, v) = 0
8
<
:
ˆ
tfl
(t, v) + ˆ
v((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
“fl
(0, v) = 0
8
<
:
ˆ
tfl
(t, V ) + ˆ
V(g(t, V )fl(t, V )) = 0
g
(t, V
0)fl(t, V
0) = —(V
p(t))
fl
(0, v) = fl
08
<
:
ˆ
tfl
(t, V ) + ˆ
V(g(t, V )fl(t, V )) = 0 t œ]0, +Œ[, V œ]V
0,
+Œ[
g
(t, V
0)fl(t, V
0) = µV
p(t)
t
œ]0, +Œ[
fl
(0, v) = fl
0V
œ]V
0,
+Œ[
8
<
:
ˆ
tfl
(t, V ) + ˆ
V(
g
(t, V )
fl
(t, V )) = 0 t œ]0, +Œ[, V œ]V
0,
+Œ[
g
(t, V
0)fl(t, V
0) =
—
(V
p(t))
t
œ]0, +Œ[
fl
(0, v) = fl
0V
œ]V
0,
+Œ[
8
<
:
ˆ
tfl
(t, v) + ˆ
v(g(v)fl(t, v)) = 0
g
(V
0)fl(t, V
0) = d(V
p(t))
fl
(0, v) = 0
ˆ
tfl
(t, v) + ˆ
v(
g
(v)
fl
(t, v)) = 0
8
>
<
>
:
ˆ
tfl
(t, v) + ˆ
v(
g
(v)
fl
(t, v)) = 0
g
(V
0)fl(t, V
0) =
d
(V
p(t))
Ä
+
s
V+Œ0d
(v)fl(t, v)dv
ä
fl
(0, v) = fl
0g
(V
0)fl(t, V
0) =
—
(V
p(t))
+
⁄
+Œ V0—
(V )
fl
(t, V ) dV
g
(V
0)fl(t, V
0) =
d
(V
p(t))
=
µ
V
p(t)
g
(V
0)fl(t, V
0) =
d
(V
p(t))
Å
+
⁄
+Œ V0d
(v)fl(t, v)dv
ã
4
Clinical problem
Early-stage breast cancer
•
94% of cases are local or regional at diagnosis but
30% will relapse
•
Estimation of the metastatic risk is key to
individualize adjuvant
therapy
•
Reduce the number of chemo cycles for patients with low risk
Brain metastases in non-small cell lung cancer (NSCLC)
•
Decide whether to use
whole brain radiation therapy
or just
Objectives
• Use a
mechanistic model
to predict metastasis
• Combine with machine learning algorithm to
select features
Clinical problem
8
<
:
ˆ
tfl
(t, v) + ˆ
v(
g(t, v)
fl
(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
fl
(0, v) = 0
g
(v)
= (
–
≠
—
ln(v)) v
8
<
:
ˆ
tfl
(t, v) + ˆ
v((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
fl
(0, v) = 0
8
<
:
ˆ
tfl
(t, v) + ˆ
v((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
“fl
(0, v) = 0
8
<
:
ˆ
tfl
(t, V ) + ˆ
V(g(t, V )fl(t, V )) = 0
g
(t, V
0)fl(t, V
0) = —(V
p(t))
fl
(0, v) = fl
08
<
:
ˆ
tfl
(t, V ) + ˆ
V(g(t, V )fl(t, V )) = 0 t œ]0, +Œ[, V œ]V
0,
+Œ[
g
(t, V
0)fl(t, V
0) = µV
p(t)
t
œ]0, +Œ[
fl
(0, v) = fl
0V
œ]V
0,
+Œ[
8
<
:
ˆ
tfl
(t, V ) + ˆ
V(
g
(t, V )
fl
(t, V )) = 0 t œ]0, +Œ[, V œ]V
0,
+Œ[
g
(t, V
0)fl(t, V
0) =
—
(V
p(t))
t
œ]0, +Œ[
fl
(0, v) = fl
0V
œ]V
0,
+Œ[
8
<
:
ˆ
tfl
(t, v) + ˆ
v(g(v)fl(t, v)) = 0
g
(V
0)fl(t, V
0) = d(V
p(t))
fl
(0, v) = 0
ˆ
tfl
(t, v) + ˆ
v(
g
(v)
fl
(t, v)) = 0
8
>
<
>
:
ˆ
tfl
(t, v) + ˆ
v(
g
(v)
fl
(t, v)) = 0
g
(V
0)fl(t, V
0) =
d
(V
p(t))
Ä
+
s
V+Œ0d
(v)fl(t, v)dv
ä
fl
(0, v) = fl
0g
(V
0)fl(t, V
0) =
—
(V
p(t))
+
⁄
+Œ V0—
(V )
fl
(t, V ) dV
g
(V
0)fl(t, V
0) =
d
(V
p(t))
=
µ
V
p(t)
g
(V
0)fl(t, V
0) =
d
(V
p(t))
Å
+
⁄
+Œ V0d
(v)fl(t, v)dv
ã
4
Experimental
Clinical
Time (days)
0 10 20 30 40 50 60 70
Primary tumor size (cells)
104 105 106 107 108 109 1010
Metastatic burden (cells)
104 105 106 107 108 109 1010 Orthotopic implantation Surgery (t=34) Pre-surgical PT Post-surgical MB Time (days) 0 10 20 30 40 50 60 70
Primary tumor size (cells)
104 105 106 107 108 109 1010
Metastatic burden (cells)
104 105 106 107 108 109 1010 Orthotopic implantation Surgery (t=34) Pre-surgical PT Post-surgical MB
SMARTc wet lab
collaboration with J. Ebos, Roswell Park Cancer Institute, Buffalo, NY
n >
400 animals
with different treatments
and schedules
Months 0 10 20 30 40 50 Metastases diameter (mm) 0 5 10 15 20 25 30 35 40 45 A Months 0 10 20 30 40 50Primary tumor diameter (mm)
0 5 10 15 20 25 30 35 40 45 TKI CT1 mTKI CT2 TGI fit Data B Months 0 10 20 30 40 50
Number of visible metastases
0 2 4 6 8 10 12 14 16 18 20 C Months 0 10 20 30 40 50 Metastases diameter (mm) 0 5 10 15 20 25 30 35 40 45 Data
Gompertz growth prediction
D
1
• Number and sizes
of brain metastasis in individual NSCLC
patients (n=31)
• Databases of metastatic relapse in breast cancer patients with
no adjuvant therapy (n=642, p=21 and n=167, p=9)
n
p
Clinical problem
8
<
:
ˆ
tfl
(t, v) + ˆ
v(
g(t, v)
fl
(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
fl
(0, v) = 0
g
(v)
= (
–
≠
—
ln(v)) v
8
<
:
ˆ
tfl
(t, v) + ˆ
v((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
fl
(0, v) = 0
8
<
:
ˆ
tfl
(t, v) + ˆ
v((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0)fl(t, V
0) =
µ
V
p(t)
“fl
(0, v) = 0
8
<
:
ˆ
tfl
(t, V ) + ˆ
V(g(t, V )fl(t, V )) = 0
g
(t, V
0)fl(t, V
0) = —(V
p(t))
fl
(0, v) = fl
08
<
:
ˆ
tfl
(t, V ) + ˆ
V(g(t, V )fl(t, V )) = 0 t œ]0, +Œ[, V œ]V
0,
+Œ[
g
(t, V
0)fl(t, V
0) = µV
p(t)
t
œ]0, +Œ[
fl
(0, v) = fl
0V
œ]V
0,
+Œ[
8
<
:
ˆ
tfl
(t, V ) + ˆ
V(
g
(t, V )
fl
(t, V )) = 0 t œ]0, +Œ[, V œ]V
0,
+Œ[
g
(t, V
0)fl(t, V
0) =
—
(V
p(t))
t
œ]0, +Œ[
fl
(0, v) = fl
0V
œ]V
0,
+Œ[
8
<
:
ˆ
tfl
(t, v) + ˆ
v(g(v)fl(t, v)) = 0
g
(V
0)fl(t, V
0) = d(V
p(t))
fl
(0, v) = 0
ˆ
tfl
(t, v) + ˆ
v(
g
(v)
fl
(t, v)) = 0
8
>
<
>
:
ˆ
tfl
(t, v) + ˆ
v(
g
(v)
fl
(t, v)) = 0
g
(V
0)fl(t, V
0) =
d
(V
p(t))
Ä
+
s
V+Œ0d
(v)fl(t, v)dv
ä
fl
(0, v) = fl
0g
(V
0)fl(t, V
0) =
—
(V
p(t))
+
⁄
+Œ V0—
(V )
fl
(t, V ) dV
g
(V
0)fl(t, V
0) =
d
(V
p(t))
=
µ
V
p(t)
g
(V
0)fl(t, V
0) =
d
(V
p(t))
Å
+
⁄
+Œ V0d
(v)fl(t, v)dv
ã
4
Lapse time from diagnosis 0
- t
diag Detectability threshold Dissemination: d Primary tumor growth g TTR lo g (T um o r siz e) Metastatic growth Pre-surgical history TTRV
diag gpV
visMathematical models
Growth rates of
primary
and
secondary
tumors
g
pand
g
Dissemination rate
Size distribution of the metastases ρ(t,v)
[Iwata et al., 2000]
Metastatic burden (total number of metastatic cells)
÷
i≥ N (0, Ê
2)
◊
i= ◊
pop+ —
Tx
i+ ÷
i,
÷
i≥ N (0, Ê
2)
x
iœ R
pN (t)
N
(t) =
⁄
t 0d
(V
p(s))ds = E [N(t)]
ln(µ) = ln(µ
m) + µ
‡Á,
Á
≥ N (0, 1)
⁄
+Œ xfl
(t, y)dy
M
(t) =
⁄
+Œ V0vfl
(t, v)dv =
⁄
t 0d
(V
p(t ≠ s)) V (s)ds = E
S
U
ÿ
iØ1V
i(t)
T
V
T
n1, ... , nk≠1, j:= T
n1, ... , nk≠1+ ˜
T
n1, ... , nk≠1, jM
(t) =
⁄
+Œ V0vfl
(t, v)dv =
⁄
t 0d
(V
p(t ≠ s)) V (s)ds
—
(V ) = mV
–dV
pdt
= (
–
p≠
—
pln(V
p)) V
pdV
pdt
=
g
p(V
p)
,
V
p(t = 0) = V
idV
pdt
=
g
p(t, V
p)
,
V
p(t = 0) = V
idV
pdt
= g
p(t, V
p)
3
8
<
:
ˆ
t
fl
(t, v) + ˆ
v
(
g(t, v)
fl
(t, v)) = 0
g
(V
0
)fl(t, V
0
) =
µ
V
p
(t)
fl
(0, v) = 0
g
(v)
= (
–
≠
—
ln(v)) v
8
<
:
ˆ
t
fl
(t, v) + ˆ
v
((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0
)fl(t, V
0
) =
µ
V
p
(t)
fl
(0, v) = 0
8
<
:
ˆ
t
fl
(t, v) + ˆ
v
((
–
≠
—
ln(v)) vfl(t, v)) = 0
g
(V
0
)fl(t, V
0
) =
µ
V
p
(t)
“
fl
(0, v) = 0
8
<
:
ˆ
t
fl
(t, V ) + ˆ
V
(g(t, V )fl(t, V )) = 0
g
(t, V
0
)fl(t, V
0
) = —(V
p
(t))
fl
(0, v) = fl
0
8
<
:
ˆ
t
fl
(t, V ) + ˆ
V
(g(t, V )fl(t, V )) = 0 t œ]0, +Œ[, V œ]V
0
,
+Œ[
g
(t, V
0
)fl(t, V
0
) = µV
p
(t)
t
œ]0, +Œ[
fl
(0, v) = fl
0
V
œ]V
0
,
+Œ[
8
<
:
ˆ
t
fl
(t, V ) + ˆ
V
(
g
(t, V )
fl
(t, V )) = 0 t œ]0, +Œ[, V œ]V
0
,
+Œ[
g
(t, V
0
)fl(t, V
0
) =
—
(V
p
(t))
t
œ]0, +Œ[
fl
(0, v) = fl
0
V
œ]V
0
,
+Œ[
8
<
:
ˆ
t
fl
(t, v) + ˆ
v
(g(v)fl(t, v)) = 0
g
(V
0
)fl(t, V
0
) = d(V
p
(t))
fl
(0, v) = 0
ˆ
t
fl
(t, v) + ˆ
v
(
g
(v)
fl
(t, v)) = 0
8
>
<
>
:
ˆ
t
fl
(t, v) + ˆ
v
(
g
(v)
fl
(t, v)) = 0
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
Ä
+
s
V
+Œ
0d
(v)fl(t, v)dv
ä
fl
(0, v) = fl
0
g
(V
0
)fl(t, V
0
) =
—
(V
p
(t))
+
⁄
+Œ
V
0—
(V )
fl
(t, V ) dV
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
=
µ
V
p
(t)
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
Å
+
⁄
+Œ
V
0d
(v)fl(t, v)dv
ã
4
ex: Gompertz
Number of metastases with size larger than the
visible size V
visN
vis(t) =
⁄
+Œ Vvisfl
(t, v)dv =
⁄
t≠·vis 0d
(V
p(t))dt
N
vis(t) =
⁄
+Œ Vvisfl
(t, v)dv
=
⁄
t≠·vis 0d
(V
p(t))dt
N
vis(t) =
⁄
+Œ Vvisfl
(t, v)dv
=
⁄
t≠·vis 0µV
p(t)dt
M
(t) = µ
⁄
t 0V
p(t)V (t ≠ s)ds
µ
i= µ
pop+ —
Tx
i+ ÷
i,
÷
i≥ N (0, Ê
2)
ln
µ
i= ln (µ
pop) + —
µTx
iµ+ ÷
µi,
÷
µi≥ N (0, Ê
µ2)
ln
–
i= ln (–
pop) + —
–Tx
–i+ ÷
–i,
÷
–i≥ N (0, Ê
–2)
–
i= µ
pop+ —
KiT 67Ki67
i+ ÷
i,
÷
i≥ N (0, Ê
2)
µ
i= µ
pop+ —
TKi67
i+ ÷
i,
÷
i≥ N (0, Ê
2)
◊
i= ◊
pop+ —
Tx
i+ ÷
i,
÷
i≥ N (0, Ê
2)
x
iln
◊
i= ln (◊
pop) + —
Tx
i+ ÷
i,
÷
i≥ N (0, Ê
2)
◊
i= ◊
pop+ ÷
i,
÷
i≥ N (0, Ê
2)
2
Time to relapse
(TTR) = time from diagnosis to first visible met
Volume (cells) 100 105 1010 1015 Number of mets 0 5 10 15 20 25 30
Imaging detection limit
g(v)
÷
i≥ N (0, Ê
2)
◊
i= ◊
pop+ —
Tx
i+ ÷
i,
÷
i≥ N (0, Ê
2)
x
iœ R
pN (t)
N
(t) =
⁄
t 0d
(V
p(s))ds = E [N (t)]
ln(µ) = ln(µ
m) + µ
‡Á,
Á
≥ N (0, 1)
⁄
+Œ xfl
(t, y)dy
M
(t) =
⁄
+Œ V0vfl
(t, v)dv =
⁄
t 0d
(V
p(t ≠ s)) V (s)ds = E
"
ÿ
iØ1V
i(t)
#
T
n1, ... , nk≠1, j:= T
n1, ... , nk≠1+ ˜
T
n1, ... , nk≠1, jM
(t) =
⁄
+Œ V0vfl
(t, v)dv =
⁄
t 0d
(V
p(t ≠ s)) V (s)ds
d
(V
p) = µV
p“—
(V ) = mV
–dV
pdt
= (
–
p≠
—
pln(V
p)) V
pdV
pdt
=
g
p(V
p),
V
p(t = 0) = V
idV
pdt
=
g
p(t, V
p),
V
p(t = 0) = V
i3
•
Classical approach considers each
subject independently
Statistical procedure for model calibration: nonlinear mixed effects modeling
Subject 1 ≤ i ≤N, Time t
jLikelihood maximization
Lavielle,CRC press, 2014g
(V
0
)fl(t, V
0
) =
—
(V
p
(t))
+
⁄
+Œ
V
0—
(V )
fl
(t, V ) dV
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
=
µ
V
p
(t)
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
Å
+
⁄
+Œ
V
0d
(v)fl(t, v)dv
ã
ˆ
t
fl
(t, V ) + ˆ
V
(
g
(t, V )
fl
(t, V )) = 0
fl
(t, V ) ≥ e
⁄
0t
(V )
⁄
+Œ
V
0fl
(t, V )dV
⁄
+Œ
V
0V fl
(t, V )dV
Y
i
j
= M(t
j
i
,
—
j
) + Á
j
i
Y
i
j
= M(t
j
i
,
—
j
) + Á
j
i
—
1
, . . . , —
N
≥
LN (—
µ
, —
Ê
)
,
—
µ
œ R
p
, —
Ê
œ R
p
◊p
y
j
i
= M(t
i
j
,
◊
i
) + ‡Á
i
j
,
Á
i
j
≥ N (0, 1)
M
j
i
= M(t
i
j
,
◊
i
) + ‡(M(t
j
i
,
◊
i
))Á
i
j
,
Á
i
j
≥ N (0, 1)
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
y
j
i
= M(t
j
i
,
◊
j
) + Á
j
i
,
◊
1
, . . . , ◊
N
≥
LN (◊
µ
, ◊
Ê
)
,
◊
µ
œ R
p
, ◊
Ê
œ R
p
◊p
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
◊
1
, . . . , ◊
N
≥
LN (◊
µ
, ◊
Ê
)
,
◊
µ
œ R
p
, ◊
Ê
œ R
p
◊p
ˆ◊
j
= min
◊
jÿ
Ä
y
i
j
≠ M(t
i
, ◊
j
)
ä
2
P(Mets) = P
Å
µ
⁄
T
10
V
(t)dt > 1
ã
5
g(V
0
)fl(t, V
0
) =
—
(V
p
(t))
+
⁄
+Œ
V
0—
(V )
fl(t, V ) dV
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
=
µ
V
p(t)
g(V
0
)fl(t, V
0
) =
d
(V
p
(t))
Å
+
⁄
+Œ
V
0d(v)fl(t, v)dv
ã
ˆtfl(t, V ) + ˆV
(
g
(t, V )
fl(t, V )) = 0
fl
(t, V ) ≥ e
⁄
0t
(V )
⁄
+Œ
V
0fl(t, V )dV
⁄
+Œ
V
0V fl(t, V )dV
Y
i
j
= M(t
j
i
,
—
j
) + Á
j
i
Y
i
j
= M(t
j
i
,
—
j
) + Á
j
i
—
1
, . . . , —
N
≥
LN (—
µ
, —
Ê
)
,
—µ
œ R
p
, —Ê
œ R
p
◊p
y
j
i
= M(t
i
j
,
◊
i
) + ‡Á
i
j
,
Á
i
j
≥ N (0, 1)
M
j
i
= M(t
i
j
,
◊
i
) + ‡(M(t
j
i
,
◊
i
))Á
i
j
,
Á
i
j
≥ N (0, 1)
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
,
◊
1
, . . . , ◊
N
≥
LN (◊
µ
, ◊
Ê
)
,
◊
µ
œ R
p
, ◊
Ê
œ R
p
◊p
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
◊
1
, . . . , ◊
N
≥
LN (◊
µ
, ◊
Ê
)
,
׵
œ R
p
, ◊Ê
œ R
p
◊p
ˆ◊
i
= min
◊
iÿ
y
j
i
≠ M(ti
, ◊
i
)
2
ˆ◊
j
= min
◊
jÿ
Ä
y
i
j
≠ M(t
i
, ◊
j
)
ä
2
5
• Population
approach
Nvis(t) =
⁄
+Œ
V
visfl(t, v)dv =
⁄
t
≠·
vis0
d(Vp(t))dt
Nvis
(t) =
⁄
+Œ
V
visfl(t, v)dv
=
⁄
t
≠·
vis0
d(Vp(t))dt
N
vis
(t) =
⁄
+Œ
V
visfl(t, v)dv
=
⁄
t
≠·
vis0
µVp(t)dt
M
(t) = µ
⁄
t
0
Vp
(t)V (t ≠ s)ds
µ
i
= µ
pop
+ —
T
x
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
ln
µ
i
= ln (µpop) + —
µ
T
x
i
µ
+ ÷
µ
i
,
÷
µ
i
≥ N (0, Ê
µ
2
)
ln
–
i
= ln (–pop) + —
–
T
x
i
–
+ ÷
–
i
,
÷
–
i
≥ N (0, Ê
–
2
)
–
i
= µ
pop
+ —
Ki
T
67
Ki67
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
µ
i
= µ
pop
+ —
T
Ki67
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
◊
i
= ◊pop
+ —
T
x
i
+ ÷i
,
÷
i
≥ N (0, Ê
2
)
x
i
ln
◊
i
= ln (◊pop) + —
T
x
i
+ ÷i
,
÷i
≥ N (0, Ê
2
)
ln ◊
i
= ln (◊
pop
) + ÷
i
,
÷
i
≥ N (0, Ê
2
)
2
+
• Parameters to be estimated = 𝜃
!"!
(𝑝), 𝜔
!(!$%)
'
and 𝜎
g
(V
0
)fl(t, V
0
) =
—
(V
p
(t))
+
⁄
+Œ
V
0—
(V )
fl
(t, V ) dV
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
=
µ
V
p
(t)
g
(V
0
)fl(t, V
0
) =
d
(V
p
(t))
Å
+
⁄
+Œ
V
0d
(v)fl(t, v)dv
ã
ˆ
t
fl
(t, V ) + ˆ
V
(
g
(t, V )
fl
(t, V )) = 0
fl
(t, V ) ≥ e
⁄
0t
(V )
⁄
+Œ
V
0fl
(t, V )dV
⁄
+Œ
V
0V fl
(t, V )dV
Y
i
j
= M(t
j
i
,
—
j
) + Á
j
i
Y
i
j
= M(t
j
i
,
—
j
) + Á
j
i
—
1
, . . . , —
N
≥
LN (—
µ
, —
Ê
)
,
—
µ
œ R
p
, —
Ê
œ R
p
◊p
y
j
i
= M(t
i
j
,
◊
i
) + ‡Á
i
j
,
Á
i
j
≥ N (0, 1)
M
j
i
= M(t
i
j
,
◊
i
) + ‡(M(t
j
i
,
◊
i
))Á
i
j
,
Á
i
j
≥ N (0, 1)
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
,
◊
1
, . . . , ◊
N
≥
LN (◊
µ
, ◊
Ê
)
,
◊
µ
œ R
p
, ◊
Ê
œ R
p
◊p
y
i
j
= M(t
j
i
,
◊
j
) + Á
j
i
◊
1
, . . . , ◊
N
≥
LN (◊
µ
, ◊
Ê
)
,
◊
µ
œ R
p
, ◊
Ê
œ R
p
◊p
ˆ◊
j
= min
◊
jÿ
Ä
y
i
j
≠ M(t
i
, ◊
j
)
ä
2
P(Mets) = P
Å
µ
⁄
T
10
V
(t)dt > 1
ã
5
Validation on animal data
Nonlinear
mixed-effects
statistical model for
inter-animal variability
Benzekry et al. (Ebos), Cancer Res, 2016
⟹
same growth
for PT and mets:
α
p
=
α
,
β
p
=
β
Ebos lab Roswell Park Cancer Institute
N
vis
(t) =
⁄
+Œ
V
visfl
(t, v)dv =
⁄
t
≠·
vis0
d
(V
p
(t))dt
N
vis
(t) =
⁄
+Œ
V
visfl
(t, v)dv
=
⁄
t
≠·
vis0
d
(V
p
(t))dt
N
vis
(t) =
⁄
+Œ
V
visfl
(t, v)dv
=
⁄
t
≠·
vis0
µV
p
(t)dt
M
(t) = µ
⁄
t
0
V
p
(t)V (t ≠ s)ds
µ
i
= µ
pop
+ —
T
x
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
ln
µ
i
= ln (µ
pop
) + —
µ
T
x
i
µ
+ ÷
µ
i
,
÷
µ
i
≥ N (0, Ê
2
µ
)
ln
–
i
= ln (–
pop
) + —
–
T
x
–
i
+ ÷
–
i
,
÷
–
i
≥ N (0, Ê
–
2
)
–
i
= µ
pop
+ —
Ki
T
67
Ki67
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
µ
i
= µ
pop
+ —
T
Ki67
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
◊
i
= ◊
pop
+ —
T
x
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
x
i
ln
◊
i
= ln (◊
pop
) + —
T
x
i
+ ÷
i
,
÷
i
≥ N (0, Ê
2
)
ln ◊
i
= ln (◊
pop
) + ÷
i
,
÷
i
≥ N (0, Ê
2
)
2
•
Main difficulty
: PDE model
•
Solution
: fast Fourier transform for
computation of metastatic burden
Clinical data of individual breast metastatic relapse
Institut Bergonié, Bordeaux, FR
K =
21
features
n
=
642
pa
tie
nt
s
w
/o
a
dj
uva
nt
outcome
Descriptive power: fit to the data
Nicolò et al., JCO: Clin Cancer Inform, 2020
improve these metrics, a balanced, downsampled version
of the data set was constructed. This significantly improved
positive predictive value and F1 as well as model calibration
(Appendix
Figure A5E-F
). We also compared our results
with classic Cox regression survival analysis, which was
found to exhibit similar predictive power (
Table 2
; Appendix
Figure A6
; Appendix
Table A2
).
DISCUSSION
We propose a mechanistic model for predicting metastatic
relapse after surgical intervention in patients with
early-stage breast cancer which, for the first time, can simulate
the pre- and postdiagnosis history of the disease from data
available at diagnosis only. Notably, with a c-index of 0.65
(95% CI, 0.60 to 0.71), the mechanistic model achieved
performances similar to those of the RSF algorithm (95%
CI, 0.66 to 0.69) and Cox regression (95% CI, 0.62 to 0.67).
These were also comparable with the actual standard
prognosis model of breast cancer–specific survival
Adju-vant! (c-index, 0.71),
7used for risk classification in the
MINDACT (ClinicalTrials.gov identifier:
NCT00433589
)
trial.
39Others also reported a similar c-index of 0.67 for
prediction of relapse in untreated patients by using Cox
analysis.
40Advanced deep learning algorithms did not
outperform this predictive power (c-index, 0.68) for
pre-diction of survival, even though they integrated genomic
data.
13A recent study that considered recurrence (local,
regional, or distant) reported superior predictive power
(area under the curve, 0.81 for prediction of relapse
at 5 years v 0.73 in our analysis), which might be
explained by the much larger data set (15,314 patients)
and inclusion of epidemiologic data not available in our
analysis.
Grounded in the biology of the metastatic process, our
model provides insights not achievable by statistical
analysis alone. First, one of the most important and
well-known predictors of metastatic relapse—tumor size
23—is
directly incorporated into the model as an input parameter
used for calculation of the tumor age. Second, our model
allows testing whether covariates are associated with
Metastasis–Free Survival
(probability)
1.0 0.8 0.6 0 5 10 15 20 Kaplan−Meier estimate Model fitTime Since Diagnosis (years)
A
Observed Probabilities
1.0 0.9 0.8 0.7 0.6 0.5 1.0 0.9 0.8 0.7 0.6 0.5Predicted Probabilities
Calibration for 2-Year Outcome
Observed Probabilities
1.0 0.9 0.8 0.7 0.6 0.5 1.0 0.9 0.8 0.7 0.6 0.5Predicted Probabilities
Calibration for 5-Year Outcome
Observed Probabilities
1.0 0.9 0.8 0.7 0.6 0.5 1.0 0.9 0.8 0.7 0.6 0.5Predicted Probabilities
Calibration for 10-Year Outcome
B
FIG 3. Fit of the mechanistic model. (A) Mean of the conditional metastasis-free survival functions from the mechanistic model and Kaplan-Meier estimates with 95% CIs (shaded areas). (B) Calibration plots of metastasis-free survival for the final mechanistic model with covariates. Vertical bars are 95% CIs.
Mechanistic Modeling of Metastatic Relapse in Breast Cancer
JCO Clinical Cancer Informatics 265
Downloaded from ascopubs.org by University of Massachusetts on March 30, 2020 from 128.119.168.112 Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
Metastasis-Free Survival
1.0
0.8
0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 10 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 20 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 15 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10
5 0
Vdiag = 25 mm
FIG A3. Distant metastasis-free survival predictions of the mechanistic model and Kaplan-Meier estimates with 95% CIs stratified by values of the primary tumor size (diameter) at diagnosis.
Mechanistic Modeling of Metastatic Relapse in Breast Cancer
JCO Clinical Cancer Informatics 271
Downloaded from ascopubs.org by University of Massachusetts on March 30, 2020 from 128.119.168.112 Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
Metastasis-Free Survival
1.0
0.8
0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 10 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 20 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 15 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10
5 0
Vdiag = 25 mm
FIG A3. Distant metastasis-free survival predictions of the mechanistic model and Kaplan-Meier estimates with 95% CIs stratified by values of the primary tumor size (diameter) at diagnosis.
Mechanistic Modeling of Metastatic Relapse in Breast Cancer
JCO Clinical Cancer Informatics 271
Downloaded from ascopubs.org by University of Massachusetts on March 30, 2020 from 128.119.168.112 Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
Metastasis-Free Survival
1.0
0.8
0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 10 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 20 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 15 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10
5 0
Vdiag = 25 mm
FIG A3. Distant metastasis-free survival predictions of the mechanistic model and Kaplan-Meier estimates with 95% CIs stratified by values of the primary tumor size (diameter) at diagnosis.
Mechanistic Modeling of Metastatic Relapse in Breast Cancer
JCO Clinical Cancer Informatics 271
Downloaded from ascopubs.org by University of Massachusetts on March 30, 2020 from 128.119.168.112 Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
Metastasis-Free Survival
1.0
0.8
0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 10 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 20 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10 5 0 Vdiag = 15 mm Metastasis-Free Survival 1.0 0.8 0.6
Time Since Diagnosis (years) 15 10
5 0
Vdiag = 25 mm
FIG A3. Distant metastasis-free survival predictions of the mechanistic model and Kaplan-Meier estimates with 95% CIs stratified by values of the primary tumor size (diameter) at diagnosis.
Mechanistic Modeling of Metastatic Relapse in Breast Cancer
JCO Clinical Cancer Informatics 271
Downloaded from ascopubs.org by University of Massachusetts on March 30, 2020 from 128.119.168.112 Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
the model remained within the Kaplan-Meier 95% CIs in all
cases, we observed that model predictions tended to be less
accurate as the tumor size increased, with overestimation of
the relapse risk for large PT sizes.
Mechanistic Covariate Analysis and Predictive Power of
the Mathematical Model
We next tested the covariates selected by the RSF analysis
in the mechanistic model. We built the covariate model by
using a backward elimination procedure, starting with the
full model with all the preselected covariates on both
pa-rameters α and µ. The final model included Ki67 and CD44
on α, and EGFR on the dissemination parameter µ as
significant covariates (
Table 1
). The cross-validated c-index
for this model was 0.65 (95% CI, 0.60 to 0.71). Calibration
plots for 2-, 5-, and 10-year outcomes demonstrated good
predictive accuracy of the model (
Fig 3B
). Nevertheless, as
in the RSF model, risk of relapse was overestimated in
high-risk groups. For the classification task of predicting 5- and
10-year relapse, comparable results were obtained
be-tween the mechanistic model, the RSF, classification
machine learning algorithms, and Cox regression (
Table 2
).
Predictive Simulations of the Mechanistic Model
The previous results allow calibration of the mechanistic
model parameters from variables available at diagnosis by
using only the covariate part in equation (2) and neglecting the
remaining unexplained variance. We used this to simulate the
natural cancer history for a number of representative
pa-tients in our data set. For each patient, the population-level
parameters α
pop, µ
pop, β
Ki67,α, β
CD44,α,and β
EGFR,µwere
calibrated from an independent training set that did not
contain this patient (coming from the cross-validation
procedure). Simulations were then performed by using
a discrete version of the metastatic model
22and are
re-ported in
Figure 4
and Appendix
Figure A4
. Covariate
values used for prediction and the resulting inferred
individual parameters ˆα
iand ˆµ
iare reported in Appendix
Table A1
.
For patients 224 and 358 (
Fig 4A-B
), the model predicted
a time from the first cell to detection (presurgical history) of
17.3 years and 3.61 years, respectively. In addition, the
model allowed prediction of the metastatic size distribution
at diagnosis (
Fig 4C-D
). Patient 224 was predicted to have
21 invisible metastases at the time of diagnosis. The largest
metastasis contained 2.32
× 10
4cells and the smallest
metastasis contained only one cell. The largest metastasis
was predicted to have initiated 12.9 years after the first cell
of the PT (ie, 4.4 years before surgery). The model predicts
relapse 6.18 years after diagnosis, whereas true relapse
occurred at 4.88 years. At diagnosis, patient 358 was
predicted to have a total of 35 invisible metastases. The
largest metastasis contained 1.16
× 10
4cells and the
smallest contained 1.36 cells. According to this simulation,
the first metastasis in this patient was emitted 2.6 years after
the PT onset (ie, 1.01 years before surgery). The model
predicted relapse at 1.97 years after diagnosis, but true
re-lapse occurred at 3.06 years. The differences in PT and
metastatic dynamics for these two patients are a result of the
different values of the covariates (Appendix
Table A1
). Distinct
levels of Ki67 cause distinct growth kinetics. Moreover, unlike
the tumor of patient 224, the tumor of patient 358 expresses
EGFR, which is associated with a higher metastatic potential.
Thus, although the tumor from patient 358 is much younger,
the total number of (invisible) metastases at surgery is
pre-dicted to be larger (35 v 21 metastases in patient 224). Model
predictions are also informative in the case of individuals who
were censored at the last follow-up. For instance, patient 70
(Appendix
Figure A4
) was censored at 17.7 years after
di-agnosis. This is consistent with our model, which predicts that
this patient was disease-free after PT resection and would
never have relapsed (TTR = +∞).
Comparison With Machine Learning Classification
Algorithms and Cox Regression
We tested the predictive power of machine learning
clas-sification algorithms. These cannot account for
right-censored data. Thus, for this part, we focused on
pre-diction of 5-year metastatic relapse (yes or no). Best
per-formances were achieved by the random forest and logistic
regression models (
Table 2
and Appendix
Figure A5A-D
).
However, owing to the low event rate (9.25%), positive
predictive value and F1 scores were low (
Table 2
). To
TABLE 1.
Parameter EstimatesParameter Estimate Relative Standard Error (%) P Model without covariates
log αpop −6.34 12.6
log µpop −26.8 3.68
σ 0.542 28.4
ωα 3.37 36.4
ωµ 3.78 15.9
Model with covariates
log αpop −9.01 10.8 βKi67,α 0.093 29.6 .001 βCD44,α 0.017 57.7 .083 log µpop −25.9 4.4 βEGFR,µ 0.053 38.1 .009 σ 0.606 24 ωα 2.75 22.1 ωµ 3.03 20.5
NOTE. Mechanistic mixed-effects models for the time to metastatic relapse were fitted to the data using likelihood maximization. Parameters α (growth) and µ (dissemination) were assumed to follow log-normal distributions with fixed effects (typical values) αpopand µpop, standard deviation of the random effects ωαand ωµ and standard deviation of the error model σ (see Methods). A first version was fitted without covariates and a second version included dependence of the individual parameters on individual covariates. P value refers to a Wald test for statistical significance of the covariate on the parameter value.
Nicol `o et al
264 © 2020 by American Society of Clinical Oncology
Downloaded from ascopubs.org by University of Massachusetts on March 30, 2020 from 128.119.168.112 Copyright © 2020 American Society of Clinical Oncology. All rights reserved.
Feature selection: random survival
forests
c-index = 0.69 (0.67 – 0.71)
M. Dmitrievsky, https://www.mql5.com/en/articles/3856