(G-250) Spatially offset Raman spectroscopy for subcutaneous implantable biosensor

Introduction:: Sub-track: Optics

Prostate cancer is the second most deadly cancer for men in the U.S [1]. Prostate-specific antigen (PSA) is a common indicator of prostate cancer; PSA levels increase in the circulatory system when prostate cancer is present [2,3]. Although a blood test appears to be an appropriate measure, PSA levels have a low specificity and are often caused by something other than cancer [2]. Immediately after treatment, PSA levels may be misleading in signaling recurrence because the levels in the interstitial fluid may not be directly related to the response [2]. The current clinical standard to monitor for recurrence is imaging, but we believe that monitoring biomolecules in the interstitial fluid would allow us to target a more accurate biomarker and achieve frequent results that can signal cancer recurrence at earlier and more treatable stages [2,3].

We designed a subcutaneous, implantable, surface-enhance Raman spectroscopy (SERS)-based biosensor to measure biomarkers for recurrence in the interstitial fluid. We want to use spatially offset Raman spectroscopy (SORS) to read this biosensor through the skin in a minimally invasive way. SORS offers the advantage over conventional Raman spectroscopy to isolate Raman signals from subsurface layers, otherwise unreachable, by introducing a spatial separation between the laser excitation and collection points that allows the Raman photons to scatter diffusely within the tissue [4]. A hydrogel containing SERS nanoparticles is used to enhance the signal of the nearby biomolecules, and SORS would isolate the subsurface spectra, extracting useful information of the composition of the interstitial fluid.

Materials and Methods:: The experiment involved varying the source-detector separation (SDS) of a custom built, free-space SORS system for two types of tissue-simulating phantoms (Figure 1). The first phantom used two layers made of polytetrafluoroethylene (PTFE) beneath polydimethylsiloxane (PDMS) to model an implant under the skin. The SDS values were varied (0.00–5.00 mm, increments of 0.25 mm) to find the optimal Raman signal for different subcutaneous implant depths (0.5, 1.0, 1.5, and 2.0 mm). The second phantom used biological components with the biosensor implant (hydrogel with gold nanostars conjugated with 4-mercaptobenzoic acid (4-MBA)) under porcine skin.

The laser of wavelength 830 nm was incident onto the samples. Acquired spectra were linearly unmixed to determine the contribution of each layer to the SORS signal, and we calculated a SORS ratio (bottom layer/top layer contribution) to determine the signal response to SDS and implant depth. By comparing the SORS ratio to a threshold of 1, we observed how SDS and implant depth influence the signal contribution of each layer. When the SORS ratio crossed over the value of 1, we identified that corresponding SDS value for our biosensor design as the optimal for the tradeoff between signal intensity and signal depth.

Results, Conclusions, and Discussions:: The SORS intensity was obtained as a function of SDS for each implant depth. At increased implant depths, the signal from the top layer (PDMS) decreased more rapidly with SDS, allowing for the isolation of the bottom layer (PTFE) signal. Beginning with the smallest implant depth, the PTFE consistently contributed over 60% to the signal. This contribution lessened with larger implant depths and required larger SDS values to overcome the top layer. For the largest implant depth, PTFE began under 40% contribution to the overall signal with zero SDS, but it converged with the top layer at a 50-50% point at 4.50 mm SDS, after which PTFE contributed to the majority of the spectra.

By comparing the SORS ratio to a threshold of 1, a greater value indicated that the bottom layer is dominant. While a 1.0 mm implant depth only required an SDS of 1.50 mm for the bottom to emerge as dominant, a 2.0 mm implant depth required an SDS of 4.50 mm. A linear relationship emerged between the SDS and implant depth (Figure 2). Using a biological phantom simulating the skin’s Raman and scattering, we were able to detect the implant’s Raman signal underneath using SORS.

SORS introduces a method of isolating deeper Raman signals from the dominant skin surface reflection and scattering. The larger SDS, however, compromises the intensity of the signal. There is an optimum SDS that can be used to measure the signal of the deeper layer. Our phantom model suggests a linear relationship between the SDS and the subcutaneous implant depth. Preliminary biological data suggests we are able to isolate the Raman signal of the implant underneath porcine skin using SORS.

Acknowledgements (Optional): : Funding for this project is provided in part by FCT (Fundacao para a Ciencia Tecnologia), Programa Operacional Regional do Norte –Norte 2020 (Projeto 45914 -04/SI/2019).

References (Optional): : [1] Prostate Cancer Statistics. Cancer.net. [Website] https://www.cancer.net/
[2] Pezaro C, Woo HH, Davis ID. Prostate cancer: measuring PSA. Intern. Med. J. 2013; 44(5): 433-40.
[3] Lilja H, Ulmert D, Vickers AJ. Prostate-specific antigen and prostate cancer: prediction, detection and monitoring. Nat. Rev. 2008; 8(4): 268-78.
[4] Matousek P, Clark IP, Draper ERC, Morris MD, Goodship AE, Everall N, Towrie M, Finney WF, Parker AW. Subsurface probing in diffusely scattering media using spatially offset Raman spectroscopy. SAS. 2005; 59(4): 393-400.