(260a) Determination and Understanding of Dynamic Lead Lag between in-Line NIR Tablet Press Feed Frame and Off-Line NIR Tablet Measurements

Process Analytical Technology (PAT) is crucial for the understanding, monitoring and control of (continuous) manufacturing processes and essential to move towards a Quality by Design (QbD) approach for pharmaceutical drug product development [1]. Different studies have evaluated near infrared (NIR) spectroscopy for blend potency determination inside the feed frame of a rotary tablet press and compared these results with off-line tablet potency measurements [2,3,4]. The final blend first passes the NIR probe before it undergoes additional mixing inside the feeding chamber and being filled into the dies of the tablet press. As a result, the in-line NIR response is always ahead of time and more intensive compared to the signal of the material inside the tablet cores (measured by offline analysis (e.g. NIR)). This phenomenon is defined as lead-lag. Although a time-shift or lead-lag has been observed between the in-line and off-line NIR response before [2], the effect of tablet press process parameters on the observed lead-lag was not investigated and no proposal was made on how to practically use the in-line NIR potency measurements for tablet diversion. Therefore, the aim of this study was to gain an in-depth understanding of the lead-lag between in-line NIR feed frame and off-line NIR tablet measurements and to propose an approach to support the use of in-line NIR feed frame measurements for product diversion and real-time release.

MATERIALS AND METHODS Materials

For this study, test formulations containing sodium saccharine as a surrogate API were used..

Blends containing different amounts of sodium saccharine were prepared using a three-dimensional mixer (Inversina, Bioengineering AG, Wald, Switzerland).

Methods

All experiments were performed using a Modul^TM P high speed rotary tablet press (GEA Pharma Systems, GEA Process Engineering, Halle, Belgium) equipped with ten 10 mm flat faced punches. All blends were compressed to a constant tablet weight of 375 mg. The NIR spectra were collected in reflection mode with a diode array spectrometer (SentroPAT FO, Sentronic, Dresden, Germany) positioned inside the feeding chamber of the feed frame, in very close proximity to the die filling station, thus monitoring blend potency immediately before the compression step [5]. Spectra were collected with an integration time of 7 ms, an averaging number of 200 spectra and a measurement cycle of 3s. A spectral range of 1100 – 2200 nm was used with a spectral resolution of 1 nm. The NIR spectra of the tablets were collected off-line by a Fourier transform near-infrared (FT-NIR) spectrometer (Antaris^TM II, Thermo Fisher Scientific, USA) in transmittance mode covering the range of 7501.74 to 11998.9 cm^-1, with a resolution of 8 cm^-1 and 64 scans per spectra. In-line and off-line NIR quantitative models were developed based on measurements of blends and tablets, respectively, referenced using different known sodium saccharine concentrations. Paddle and turret speeds were varied according to a full factorial design of experiments (DoE), for both model calibration and validation. The NIR models were developed using the SIMCA software (Version 15, Umetrics, Umeå, Sweden).

In order to measure the lead-lag, two series of experiments were conducted. The first series of experiments were conducted by performing blend composition step-changes (90% - 110%) of sodium saccharine during tablet compression and, the second, by adding a spike of pure sodium saccharine into the feed chute above the tablet press feed frame. To enhance the understanding of the influence of process parameters during tablet compression on the observed lead-lag; turret speed, paddle speeds, paddle speed ratio, overfill level and feed frame design were varied according to a DoE. Other equipment parameters such as paddle wheel design, position of the NIR probe and amount of punches were kept constant.

Concentration versus time profiles were plotted and lead-lag was determined.

RESULTS AND DISCUSSION

An off-line univariate linear regression NIR model using the wavelength at 8801.52 cm^-1 and an in-line PLS model using the spectral range between 1600 and 1750 nm were obtained. The two models had an RMSEP % of 0.40 and 0.44 respectively.

From the step change experiments, it was observed that the lead-lag decreases at higher paddle speeds due to the similar responses in blend potency between NIR and compressed tablets. This phenomenon can be explained by the higher mixing intensity inside the feed frame. Similarly, the lead-lag was observed to follow the same behaviour at higher paddle speeds, when conducting spike experiments. The turret speed and overfill level were found not having a significant influence on the lead-lag within the experimental process ranges studied.

An approach was developed to use current experimental findings for supporting future use of in-line NIR for in-process control (IPC), product diversion and real-time release.

REFERENCES

[1] Wahl, P. R., Fruhmann, G., Sacher, S., Straka, G., Sowinski, S., & Khinast, J. G. (2014). PAT for tableting: Inline monitoring of API and excipients via NIR spectroscopy. European Journal of Pharmaceutics and Biopharmaceutics, 87(2), 271–278.

[2] Pauli, V., Roggo, Y., Pellegatti, L., Nguyen Trung, N. Q., Elbaz, F., Ensslin, S., ... Krumme, M. (2019). Process analytical technology for continuous manufacturing tableting processing: A case study. Journal of Pharmaceutical and Biomedical Analysis, 162, 101–111.

[3] Sierra-Vega, N. O., Román-Ospino, A., Scicolone, J., Muzzio, F. J., Romañach, R. J., & Méndez, R. (2019). Assessment of blend uniformity in a continuous tablet manufacturing process. International Journal of Pharmaceutics, 560, 322–333.

[4] Šašić, S., Blackwood, D., Liu, A., Ward, H. W., & Clarke, H. (2015). Detailed analysis of the online near-infrared spectra of pharmaceutical blend in a rotary tablet press feed frame. Journal of Pharmaceutical and Biomedical Analysis, 103, 73–79.

[5] De Leersnyder, F., Peeters, E., Djalabi, H., Vanhoorne, V., Van Snick, B., Hong, K., ... De Beer, T. (2018). Development and validation of an in-line NIR spectroscopic method for continuous blend potency determination in the feed frame of a tablet press. Journal of Pharmaceutical and Biomedical Analysis, 151, 274–283.