From: Mark Sale <*msale*>

Date: Sat, 30 May 2015 02:44:48 +0000

Thanks to Mannie, this is the correct way to solve this (and seems to work =

as well).

Thanks

Mark

From: echigutsa

Sent: Friday, May 29, 2015 7:34 PM

To: Mark Sale

Subject: Re: [NMusers] Modeling accelerated phase of malignancy

Hi Mark

Have you tried using MTIME and MPAST for the switch? In my experience it ap=

pears to behave better for time-dependent changes.

Best wishes,

Mannie

Sent from Yahoo Mail on Android<https://overview.mail.yahoo.com/mobile/?.sr=

c=Android>

________________________________

From: Mark Sale <msale

To: nmusers

nm.com<mailto:nmusers

Subject: [NMusers] Modeling accelerated phase of malignancy

Sent: Fri, May 29, 2015 7:52:36 PM

Dear Colleagues,

I'm working on a model of a malignancy that, at some point in the course =

of the disease enters into an accelerated phase. I'm using a sort of stand=

ard serial compartment model, with a zero order input rate, then first orde=

r transit to the next compartment. I think the "correct" model for natural=

history is a slow increase in the input rate over time, then, at some poin=

t change to an exponential growth. I'm having trouble getting NONMEM to do=

this. The relevant code I have in $DES is:

IF(T.LT.NTLAG) THEN

LGIND = 0

ELSE

LGIND = 1

END IF

NATHL = LGIND*NATH

DADT(3) = (INPUT-A(3)*K)+ A(3)*NATHL

DADT(4) = A(3)*K-A(4)*K

.

.

.

Where NTLAG is an estimated parameter for the lag time between entry into t=

he study and the onset of the accelerated phase, NATH is the natural histor=

y term, NATHL is the lagged natural history term, INPUT is the zero order i=

nput rate and K is the first order transit constant. FOCE actually works =

pretty well for this for the THETA term for NTLAG, gives reasonable values.=

Probably is with the ETA for NTLAG (which is essential since it varies fr=

om person to person. With FOCE I get zero gradient for it. BAYES, SAEM, I=

MP MAP and ITS give reasonable values for OMEGA, but conditional values for=

ETA are all zero.

What I think is going on is that, unlike an ALAG, there is not event at tha=

t point in time, so small changes in ETA (smaller than the integration step=

size) don't change the predicted value, so no gradient and all ETAs = 0 =

with EM methods.

I've tried to figure out a way to do this with an additional compartment fo=

r the natural history and haven't been able to yet. That, I think would so=

lve the problem, since an event would be inserted at the end of ALAG.

Any ideas on a solution, is there a way to insert an event at an unknown ti=

me?

Thanks

Mark

Received on Fri May 29 2015 - 22:44:48 EDT

Date: Sat, 30 May 2015 02:44:48 +0000

Thanks to Mannie, this is the correct way to solve this (and seems to work =

as well).

Thanks

Mark

From: echigutsa

Sent: Friday, May 29, 2015 7:34 PM

To: Mark Sale

Subject: Re: [NMusers] Modeling accelerated phase of malignancy

Hi Mark

Have you tried using MTIME and MPAST for the switch? In my experience it ap=

pears to behave better for time-dependent changes.

Best wishes,

Mannie

Sent from Yahoo Mail on Android<https://overview.mail.yahoo.com/mobile/?.sr=

c=Android>

________________________________

From: Mark Sale <msale

To: nmusers

nm.com<mailto:nmusers

Subject: [NMusers] Modeling accelerated phase of malignancy

Sent: Fri, May 29, 2015 7:52:36 PM

Dear Colleagues,

I'm working on a model of a malignancy that, at some point in the course =

of the disease enters into an accelerated phase. I'm using a sort of stand=

ard serial compartment model, with a zero order input rate, then first orde=

r transit to the next compartment. I think the "correct" model for natural=

history is a slow increase in the input rate over time, then, at some poin=

t change to an exponential growth. I'm having trouble getting NONMEM to do=

this. The relevant code I have in $DES is:

IF(T.LT.NTLAG) THEN

LGIND = 0

ELSE

LGIND = 1

END IF

NATHL = LGIND*NATH

DADT(3) = (INPUT-A(3)*K)+ A(3)*NATHL

DADT(4) = A(3)*K-A(4)*K

.

.

.

Where NTLAG is an estimated parameter for the lag time between entry into t=

he study and the onset of the accelerated phase, NATH is the natural histor=

y term, NATHL is the lagged natural history term, INPUT is the zero order i=

nput rate and K is the first order transit constant. FOCE actually works =

pretty well for this for the THETA term for NTLAG, gives reasonable values.=

Probably is with the ETA for NTLAG (which is essential since it varies fr=

om person to person. With FOCE I get zero gradient for it. BAYES, SAEM, I=

MP MAP and ITS give reasonable values for OMEGA, but conditional values for=

ETA are all zero.

What I think is going on is that, unlike an ALAG, there is not event at tha=

t point in time, so small changes in ETA (smaller than the integration step=

size) don't change the predicted value, so no gradient and all ETAs = 0 =

with EM methods.

I've tried to figure out a way to do this with an additional compartment fo=

r the natural history and haven't been able to yet. That, I think would so=

lve the problem, since an event would be inserted at the end of ALAG.

Any ideas on a solution, is there a way to insert an event at an unknown ti=

me?

Thanks

Mark

Received on Fri May 29 2015 - 22:44:48 EDT