• Magazine
  • ENEWS
  • SUBSCRIBE
  • WEBINARS
  • PODCASTS
  • EBOOKS
    1. Home

    Ciurczak’s Response on PAT, Potato Chips and Dissolution Testing

    Aug. 24, 2006
    Emil Ciurczak responds to Frank Sistare's comments on his article. Any comments, anyone? "...I am "cautious" only in that I do not want people to think that PAT is easy or cheap to make work. My early (and current) experiences with NIR are that the salesmen ONLY discuss the application of NIR (fast, simple to operate, nondestructive) and gloss over the (possibly) months to a year of development and validation required for a method to be ready for "prime time." This could slow PAT development and innovation in the pharmaceutical manufacturing environment. When any method in a compendial journal (USP, EP, etc.) has troubles, it will show up as trouble. That is, an interference in, say, a Karl Fischer titration will give pretty radical results. A dissolution will show the release rate, no matter what is causing it. With an NIR prediction, however, you CANNOT possibly a priori know everthing that will change over the lifetime of a product's manufature. Since everything affects an NIR spectrum, you could potentially fail a good product or pass a bad product, based on an equation built without a new change having been accounted for. It is difficult to fool NIR for a specific analyte, but, considering the complexity of diffuse reflection/transmission (for which no theoretical explanation has been seen, to date), many things could give a false reading.        I just completed teaching another 3-day NIR course and some of these things were addressed. I have been doing dissolution predictions since 1989 and have seen as many bogus examples as successful ones. NO equation is static; all predictive methods must be routinely challanged. Do I think that we should trust NIR? Of course. Do we trust our kids? Of course; but we have them phone us and tell us where they are and when they're coming home, anyway. PAT is a TOOL. It is boundless. And it is limited by our imagination. But, will we ever throw away the compendial methods upon which we built PAT? I think the jury is still out on that. I think PAT is inevitable and will be the only means American (and other large) Pharma houses can stave off low cost products from developing countries and the threat of government price controls. It is smart, logical, and will result in better products..." -AMS  
    Emil Ciurczak responds to Frank Sistare's comments on his article. Any comments, anyone? "...I am "cautious" only in that I do not want people to think that PAT is easy or cheap to make work. My early (and current) experiences with NIR are that the salesmen ONLY discuss the application of NIR (fast, simple to operate, nondestructive) and gloss over the (possibly) months to a year of development and validation required for a method to be ready for "prime time." This could slow PAT development and innovation in the pharmaceutical manufacturing environment. When any method in a compendial journal (USP, EP, etc.) has troubles, it will show up as trouble. That is, an interference in, say, a Karl Fischer titration will give pretty radical results. A dissolution will show the release rate, no matter what is causing it. With an NIR prediction, however, you CANNOT possibly a priori know everthing that will change over the lifetime of a product's manufature. Since everything affects an NIR spectrum, you could potentially fail a good product or pass a bad product, based on an equation built without a new change having been accounted for. It is difficult to fool NIR for a specific analyte, but, considering the complexity of diffuse reflection/transmission (for which no theoretical explanation has been seen, to date), many things could give a false reading. I just completed teaching another 3-day NIR course and some of these things were addressed. I have been doing dissolution predictions since 1989 and have seen as many bogus examples as successful ones. NO equation is static; all predictive methods must be routinely challanged. Do I think that we should trust NIR? Of course. Do we trust our kids? Of course; but we have them phone us and tell us where they are and when they're coming home, anyway. PAT is a TOOL. It is boundless. And it is limited by our imagination. But, will we ever throw away the compendial methods upon which we built PAT? I think the jury is still out on that. I think PAT is inevitable and will be the only means American (and other large) Pharma houses can stave off low cost products from developing countries and the threat of government price controls. It is smart, logical, and will result in better products..." -AMS
    About the Author

    pharmamanufacturing | pharmamanufacturing