Photodynamic therapy (PDT) based on photodynamic reaction has been established as a novel treatment modality for cancers and precancerous lesions. PDT adopts the light with an appropriate wavelength in the presence of oxygen to activate a photosensitizer and generate reactive oxygen species (ROS), which then causes localized cell death or tumor necrosis. Concisely, PDT treatment may be described by specifying the administered photosensitizer dose, treatment light dose, and drug-light interval. As a basic component of PDT, the spatial distribution of photosensitizer may directly influence the efficacy of PDT. In vivo quantitative measurement of photosensitizer concentration provides the basis for personalized PDT. In addition to the purpose of treatment, the information on the spatial distribution of a photosensitizer in tissue can be used to identify the tumor tissue and associated margins better.

Accurate quantification of in vivo photosensitizer concentration remains challenging. In addition to the photosensitizer, fluorescence also originates from other endogenous fluorophores. Furthermore, intrinsic and instrumental factors also affect the measured fluorescence signal, limiting the ability to make accurate and reliable measurements. Hence, various technologies have been developed to quantify the fluorescence of photosensitizers. This review introduces the instrumentation and intrinsic factors impacting the in vivo quantitative detection of photosensitizer concentrations. Here, recent research progress in the fluorescence correction methods and quantitative detection techniques of photosensitizers are summarized. Finally, the potential challenges and the prospects of quantitative detection techniques of photosensitizer in the clinical translational application of PDT are also briefly discussed.

The quantitative detection of photosensitizer concentration is a complex process. The measured fluorescence intensity is influenced by instrumental factors, which include excitation light sources, optical components, detectors, and computers (Fig. 4). Moreover, the measured signal is also significantly impacted by tissue optical properties, which include the scattering and absorption of the excitation light and fluorescence emissions (Fig. 5). Consequently, several techniques have been developed to correct the measured fluorescence for endogenous fluorescence, tissue optical properties, and instrumental factors. The correction techniques can be broadly categorized as empirical, Monte Carlo (MC) simulation, and theoretical methods (Fig. 6).

The empirical methods have been commonly used to compensate for the attenuation triggered by tissue absorption and scattering on the excitation light and measured fluorescence emissions, which mainly include subtraction and ratio techniques. The empirical methods have the potential to enable near real-time data processing, owing to the inherent simplicity of the proposed methods. MC simulations are most widely used to correct fluorescence measurements, and MC modeling can be used to simulate fluorescence signals collected by an isotropic detector placed on a tissue surface with varying optical properties. The parameters for correction factors can be readily obtained from the MC simulations. The conventional theoretical methods mainly include diffusion theory, modified Beer-Lambert law, and Kubelka-Munk theory. The theoretical methods usually necessitate calculating the transfer function relating intrinsic to measured fluorescence.

To date, technologies for fluorescence quantification have used either contact, handheld spectroscopic probes, or non-contact, wide-field imaging systems (Fig. 7). The handheld spectroscopic probes have proven to be an effective technique for precisely quantifying the photosensitizer concentration by utilizing rigorous correction methods (Fig. 8). The handheld fiber-optic probe can quantitatively measure photosensitizer (e.g., chlorine e6) concentration in vivo. Consequently, handheld probes have commonly been used as the “gold standard” for fluorescence qualification. However, it can only measure a small field of view for each acquisition. Wide-field imaging systems allow imaging of the spatial distribution of photosensitizer over a larger area (Fig. 9). This technique provides a map of fast estimation of photosensitizer concentration across the field of view. Despite the aforementioned advantages, fluorescence quantification using a wide-field imaging system remains challenging because fluorescence imaging is highly sensitive to lighting variations and varying distances (e.g., distance between excitation and tissue, distance between tissue and detector).

In addition to spectroscopic probes and wide-field systems, several novel quantitative techniques have been proposed for fluorescence quantification, including fluorescence tomography, single-cell resolved microscopic system, portable imaging system, and endoscopic imaging system. Fluorescence tomography enables the 3D spatial distribution information of photosensitizer to be obtained. Single-cell resolved microscopic system is an encouraging technique for imaging tissue at cellular resolution and has the potential to reveal intra-tumor heterogeneity. Portable quantitative fluorescence imaging systems (e.g., smartphone-based systems) provide convenient image collection, computation, and quantitative imaging guidance at the point of care. The endoscopic fluorescence quantitative imaging system is intended for in vivo imaging of internal body organs.

In vivo qualification of photosensitizer concentration is crucial for personalized PDT and cancer diagnosis. However, the quantitative detection of photosensitizer concentration is a complex process. Ongoing research attempts to develop the depth-resolved, high-sensitivity, high-resolution optical imaging technique for in vivo real-time quantification of the photosensitizer concentration for pre-, during- and post-PDT.