Dear All, I was wondering whether any one on the forum can help me with fine tuning my fitting of temperature - time data. I used a IDR model I (inhibition of production). I have temperature - time data at 3 doses. Lowest dose (0.3 mg/kg) does not cause inhibition of temperature For the Middle dose (3 mg/kg), I used IDR model i and the fit looks perfect. On the other hand, at the highest dose (30 mg/kg), the lowering of body temperature reaches its physiological limits. I used IDR model with limits to fit the data. Unfortunately, I am not satisfied with the fit (not capturing the flat portion of the response - time data; kindly see Phoenix file and excel sheet). I have tried different things with no luck, I would greatly appreciate if any one on the forum could help fine tune my fitting. I would greatly appreciate if any one on the forum can guide me on how to simultaneously fit response - time data at 3 doses in Phoenix. Thank you very much in advance. [file name=Temp_data_at_3_doses.xls size=138752]Certara | Drug Development Solutions [file name=IDR_with_Limits_30_mpk.phxproj size=684184]Certara | Drug Development Solutions (668 KB)Temp_data_at_3_doses.xls (136 KB)
Dear Kris I developed a model using the population approach. the results seem not bad to me. There is variability in the model parameters that requires population approach to get individual good fits. It is attached. Look at the last model and the plot too. Let me know if it is helpful. best Serge [file name=Chaitu_Hi_Funda_Model.phxproj size=2192178]Certara | Drug Development Solutions (2.09 MB)
Dear Serge, thank you very much for your prompt response. Unfortunately, I am not able to open the attachment. I have Phoenix v6.2 and I am getting a error message saying that Phoenix used in your attachment is a newer version (v6.3). Can you kindly let me know if there is any other way to open your attachment. Thank you again for your help. Regards, kris
Hi Kris, Version 6.3 is quite old now at almost 2 years since release so I would suggest you consider planning your upgrade as soon as you can in the knowledge that 6.4 will appear later this year. I have pasted the code of the model out and attach the settings.txt you can use to confirm the mappings. test(){ deriv(E = Kin * (1 - Imax * C ^ gam / (C ^ gam + IC50 ^ gam)) - Kout * E) cfMicro(A1, Cl / V, Cl2 / V, Cl2 / V2) dosepoint(A1) C = A1 / V sequence{E = Kin / Kout} error(EEps = 0.0203543) observe(EObs = E * (1 + EEps)) stparm(V = tvV) stparm(V2 = tvV2) stparm(Cl = tvCl) stparm(Cl2 = tvCl2) stparm(Kin = tvKin * exp(nKin)) stparm(Kout = tvKout * exp(nKout)) stparm(Imax = tvImax * exp(nImax)) stparm(IC50 = tvIC50 * exp(nIC50)) stparm(gam = tvgam * exp(ngam)) fixef(tvV(freeze) = c(, 1348.9, )) fixef(tvV2(freeze) = c(, 2926, )) fixef(tvCl(freeze) = c(, 30.4734, )) fixef(tvCl2(freeze) = c(, 6.27353, )) fixef(tvKin = c(, 637.14, )) fixef(tvKout = c(, 17.582, )) fixef(tvImax = c(, 0.193169, )) fixef(tvIC50 = c(, 61.2819, )) fixef(tvgam = c(, 2.8714, )) ranef(diag(nKin, nKout, nImax, nIC50, ngam) = c(2.1950438E-07, 1.3795465E-08, 0.99575335, 0.99808717, 0.99559278)) } [file name=Settings.txt size=1928]Certara | Drug Development Solutions (1.88 KB)
Dear Serge, thank you very much for your response. I did try out your code (with pop variability). Some how I could not wrap my around what was going on (I have very limited population / statistical experience). My main question is : Can I simultaneously fit PD (temperature) data from 3 doses without coding for the population variability. Kindly see attached phoenix files for individual dose fits. I get good fitting when I model individual doses separately. The problem is that I am not able to make phoenix work for simultaneous fitting of 3 doses (kindly see attached phoenix file for individual and simultaneous fitting of 3 doses). My guess is that I am making a error in mapping the doses. I would greatly appreciate if you can help me with this. If possible, screen shots on a generic way to simultaneous fit PD data would be of great help. I followed a method in one of the threads in the forum on simultaneous fitting of PK at 2 doses and adopted that method to my case, but it did not work. [file name=IDR_3_mpk.phxproj size=667863]Certara | Drug Development Solutions (652 KB)
Dear Serge, kindly find attached phoenix files for Individual fitting at highest dose (30 mpk) and simultaneous fitting of all 3 doses (which did not work). [file name=IDR_30_mpk.phxproj size=684926]Certara | Drug Development Solutions (669 KB)
Dear Serge, kindly see attached phoenix file for simultaneous fitting at all 3 dose (which did not work). I am guessing my dose mapping may be wrong. Kindly help. [file name=IDR_All_3_doses.phxproj size=711644]Certara | Drug Development Solutions (695 KB)
Dear Kris I modified your model to have it at least running with phoenix Winnonlin. I am a population guy but at least I changed the model to enable the 3 doses. There are different ways to do it. Please take a look and come back to me if needed. With the initial parameters you out, the low dose shows a small decrease first that does not show up in the observations. This apparently is a model misspecification. Best Regards Serge [file name=IDR_All_3_doses-20140325.phxproj size=1196439]Certara | Drug Development Solutions (1.14 MB)