Quantification of the active ingredient in a pharmaceutical topical gel formulation

Importance of the topic

The quantification of active pharmaceutical ingredients in topical formulations is critical for product quality, safety, and regulatory compliance. Topical gels, creams and ointments are analytically challenging because they often require laborious extraction and sample preparation for conventional assays. Rapid, nondestructive alternatives that minimize sample handling and accelerate routine testing are therefore highly valuable for quality control and process analytical technology (PAT) applications.

Objectives and study overview

This application note evaluates Fourier transform near-infrared (FT-NIR) transmission spectroscopy as a quantitative method for determining ketoprofen concentration in a translucent carbopol gel matrix. The study prepared gels with 0, 1, 2, 4, 6 and 8 % (w/w) ketoprofen, collected transmission FT-NIR spectra, developed multivariate calibrations, and assessed method performance with independent check samples and replicate measurements.

Methodology

Gels were produced by dissolving Carbopol 980 into a water/propylene glycol mixture to create a stock gel base; ketoprofen was dissolved in a water/propylene glycol/triethanolamine cosolvent and mixed into the Carbopol stock to yield target drug levels. Prepared gels were transferred to 7-mm diameter disposable glass vials and centrifuged to remove air bubbles. Four aliquots per formulation were measured to evaluate reproducibility.

Spectral acquisition parameters:

• Instrument: Thermo Scientific Antaris FT-NIR Analyzer (transmission module)
• Spectral range: 10000–4000 cm⁻¹
• Resolution: 4 cm⁻¹
• Co-averaged scans: 64
• Typical acquisition time: ~47 seconds per spectrum
• Detector: InGaAs

Chemometric approach:

• Software: Thermo Scientific TQ Analyst
• Calibration algorithms: Stepwise Multiple Linear Regression (SMLR) and Partial Least Squares (PLS‑I)
• Preprocessing methods explored: Multiplicative Scatter Correction (MSC), Norris derivatives, Savitzky–Golay derivatives
• Final SMLR single-point model used the spectral intensity at 8792 cm⁻¹ (assigned to the second C–H overtone of ketoprofen)
• Validation: leave-one-level-out cross-validation (RMSECV) and external check samples omitted from the calibration set

Instrumention used

The measurements were performed on a Thermo Scientific Antaris FT‑NIR Analyzer in transmission mode with an InGaAs detector. The report notes the existence of an improved model (Antaris II) with enhanced speed and performance compared to the older Antaris instrument used in this study.

Main results and discussion

Ketoprofen powder produced distinct FT‑NIR features, and second‑derivative spectra of gels showed resolvable spectral differences across concentration levels sufficient for quantitative modeling. The SMLR single‑point calibration at 8792 cm⁻¹ produced excellent linearity (correlation coefficient R² ≈ 0.9996). Key error metrics were:

• RMSEC (calibration): 0.0775 % (absolute ketoprofen content)
• RMSECV (leave-one-level-out): 0.0990 %

External check samples (1.0, 4.0 and 6.0 % actual ketoprofen) were predicted with negligible bias: predicted values of 1.00, 4.02 and 5.98 % respectively (relative differences 0, +0.55, −0.33 %). Six replicate measurements of the 4 % sample yielded %RSD = 0.10 %, demonstrating high precision. These statistics indicate the FT‑NIR transmission method achieves accuracy and precision compatible with routine pharmaceutical assay requirements for this translucent gel system.

Benefits and practical applications

• Minimal sample preparation: analyses performed directly on gel-filled vials without extraction.
• Non-destructive and rapid: spectral acquisition in under a minute supports high throughput QC testing.
• High precision and low bias for the tested formulation: suitable for batch release testing, in‑process control and PAT implementations for translucent gel products.
• Reduced laboratory workload and faster turnaround versus chromatographic or extraction‑based methods.

Future trends and potential uses

• Instrument evolution and speed: newer FT‑NIR platforms (e.g., Antaris II and successors) will improve throughput and signal quality.
• Expanded chemometrics: more robust multivariate models, transfer learning and calibration transfer techniques will extend applicability across batches, instruments and related formulations with different excipient ratios or pigmentation.
• Broader PAT integration: inline or at‑line NIR sampling and automated multivariate monitoring can support process control during gel manufacture.
• Regulatory acceptance: continued demonstration of equivalence to reference assays and robust validation strategies will facilitate regulatory adoption for QC release testing.
• Advanced modalities: hyperspectral imaging or coupling NIR with other spectroscopic techniques may enable spatially resolved content uniformity assessments in semi‑solid matrices.

Conclusion

This study demonstrates that FT‑NIR transmission spectroscopy is a viable, rapid and precise method for quantifying ketoprofen in a translucent carbopol gel matrix without extensive sample preparation. The method delivered excellent linearity, low prediction errors and high repeatability for the tested formulation range (0–8 % w/w), indicating FT‑NIR can serve as an effective alternative to slower extraction‑based assays for similar topical products.

References

1. J.K. Drennen, E.G. Kramer and R.A. Lodder, Critical Reviews in Analytical Chemistry, 22(6), 443 (1991).
2. J.D. Kirsch and J.K. Drennen, Applied Spectroscopy Reviews, 30(3), 139 (1995).
3. K.M. Morisseau and C.T. Rhodes, Drug Development and Industrial Pharmacy, 21, 1071 (1995).
4. K.B. Bradfield and R.A. Forbes, Journal of Near Infrared Spectroscopy, 5, 41 (1997).
5. I.R. Last and K.A. Prebble, Journal of Pharmaceutical and Biomedical Analysis, 11 (11/12), 1071 (1993).
6. T. Isaksson and T. Næs, Applied Spectroscopy, 42, 1273 (1988).
7. P. Williams and K. Norris, Cereal Chemistry, 61, 158 (1984).
8. A. Savitzky and M.J.E. Golay, Analytical Chemistry, 36, 1627 (1964).