BioPharm

BioPharm October eBook: Best Practices 2018

Issue link: https://www.e-digitaleditions.com/i/1043363

Contents of this Issue

Navigation

Page 20 of 27

www.biopharminternational.com October 2018 BioPharm International eBook 21 instruments first came out, they were limited to a one order-of- magnitude range of concentration. Current instruments have a wider range and the laser can be tuned; samples can be measured in a non- destructive manner. Most spectroscopy techniques are not new; however, instruments have been adapted to be faster, less expensive, able to work with smaller samples, and work with a wider range of conditions. And, what is one technique's weakness is another technique's strength, Roberts says. CONCENTRATING ON RESULTS Fluorescence spectroscopy and CD spectroscopy were developed to work at very low concentration samples. A downfall of the instru- ments is if you go to 10 0 mg/ mL, you will saturate the detec- tor; so, you have to dilute, Roberts explains. A historic weakness for IR and Raman spectroscopy was that they were reliable only at high concen- trations, because the signal inher- ent ly is much wea ker, Rober ts explains. The M MS technolog y from Redshift Bio has a different instrument design but is still IR spectroscopy. "And now you have a wider dynamic working range, you can work without diluting," he says. B iophy s ic a l te c h n iqu e s c a n be limited in the concentration ranges they can run. "CD has a n a r r ow l i ne a r r a n ge you c a n work in, especially far UV," says Kendrick. "Sometimes you have to dilute your sample quite a bit to get it down to a concentration that is compatible with the mea- surement; you don't know for sure whether the dilution is tweaking your profile." A technique may require a larger volume and concentration of the sample, which is not available, forc- ing researchers to use an inferior technique—or no technique at all— which can create a gap in knowl- edge about the formulation. If sample manipulation involved is required to perform a technique, interpreting the data to under- stand how the sample is behaving becomes a challenging, subjective exercise. "It is still a little bit of a chal- lenge where there are different techniques optimal at different concentration ranges and you have to understand there could poten- tially be concentration effects that go along with that," says Yoder. "So, often times you are expected or forced to use orthogonal meth- ods to validate certain data out- puts." "In development, you really don't have a lot of material; it is in short supply, and it is very difficult to get enough sample to do the analysis," says Fleming. "It is really good to have a technique that is very sensi- tive, needs a very small amount of material, and that doesn't require manual labor to do the testing." "Techniques that allow you to study your product in its native conditions like DLS, AUC, or MMS are partic ularly usef ul in dr ug development," says Shah. Fleming notes that the MMS technology can analyze secondary structure by IR using 0.1 mg/mL concentration, providing "decent quality spectra." Yoder notes the importance of automation in light of pressures to accelerate development time- lines. He cautions, however, that automat ion ca n generate more data. "Collecting that and being able to easily sor t t hrough t he data and make, hopefully, unam- biguous assessments of that data quickly is going to be important. T he sheer amount of data gen- erated now to support filings is, a nd w i l l cont i nue, to i nc rease over time." NEW TECHNOLOGIES FOR NEW TIMELINES Training will play a more pivotal role as more sophisticated tech- niques generate more data that will require analysis and interpreta- tion, especially for multiple samples across multiple platforms and mul- tiple products, says Yoder. "I think we are going to see an increasing interest in machine learn- ing and artificial intelligence to look at why the product was successful or why was it a failure from a bio- physical or stability standpoint or efficacy standpoint," says Yoder. "I think tools will go beyond saying: 'I can generate a spectra and print that out and interpret it.' I think it will be increasingly important to under- stand a couple dozen options: What physical form of the mAb or what presentation of a mAb or biologic is going to be the most efficacious or the most chemically stable? Those are difficult questions to answer anyway without generating data, but then understanding why that data is the way it is, is very important in moving forward in development." Some analytical advances result from new applications for existing technologies that are streamlined for biological material studies and productivity requirements. "Most new platforms are based on automation with the aim of making workflows simpler, more consistent, and designed for higher throughput of samples," Shah says. Kendrick notes the importance of understanding different tools and figuring out how to applying them to techniques that are completely unrelated—like spectroscopy and pa r t ic u late a na lysis — ca n help everybody continue to improve those tools and get ideas across dif- ferent platforms. BP Biopharma Laboratory Best Practices Protein Characterization

Articles in this issue

Links on this page

Archives of this issue

view archives of BioPharm - BioPharm October eBook: Best Practices 2018