Skip to main content
NCBI home page
Biomedical Optics Express logoLink to Biomedical Optics Express
. 2025 Jun 24;16(7):2959–2971. doi: 10.1364/BOE.565339

Multi-parametric functional optical spectroscopy to monitor the metabolic and vascular changes in small head and neck tumors in vivo with radiation stress

Jing Yan 1, Pranto Soumik Saha 1, Md Zahid Hasan 1, Cristina M Furdui 2, Caigang Zhu 1,*
PMCID: PMC12265602  PMID: 40677807

Abstract

We demonstrated a portable multi-parametric functional optical spectroscopy to monitor metabolic and vascular changes in small head and neck tumors in vivo with fractional radiation therapy. For the first time, we captured the key metabolic and vascular parameters of head and neck xenograft tumors in vivo prior to and post a total of 10 Gy fractional radiation therapy. Our animal studies showed dramatic vascular and metabolic changes in radioresistant head and neck tumors (rSCC-61) under radiation stress but not in radiosensitive head and neck tumors (SCC-61). Specifically, our data showed that rSCC-61 tumors had increased tissue oxygen saturation (indicating reoxygenation), increased total hemoglobin content (indicating blood perfusion), and increased oxygenated hemoglobin (indicating oxygen supply) post radiation therapy. Our study also showed that rSCC-61 tumors had decreased glucose uptake and increased mitochondrial function post-radiation therapy. In contrast, SCC-61 tumors had minimal changes in either vascular or metabolic parameters post-radiation treatment. These results demonstrated the potential of our portable multi-parametric functional optical spectroscopy to evaluate tumor vascular and metabolic changes under therapeutic stresses for future head and neck cancer research.

1. Introduction

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer globally and remains a substantial public health challenge in the United States [1,2]. Radiotherapy (RT) alone or combined with chemotherapy is used as a primary treatment modality for over 75% of locally advanced HNSCC patients [3], while over 50% of these RT treated HNSCC patients are prone to develop tumor recurrence [3]. Radioresistance poses a critical clinical challenge, significantly limiting treatment efficacy and contributing to poor patient outcomes. Given the high recurrence rates of HNSCC following RT, a holistic understanding of the mechanisms underlying radioresistance will provide critical insights into tumor radioresistance and guide strategies to improve clinical outcomes for HNSCC patients.

Hypoxia is recognized as the major factor that undermines the RT outcome [4]. Radiation relies on oxygen to enhance DNA damage via reactive oxygen species (ROS) while hypoxic regions within tumors diminish this effect, reducing RT efficacy [5,6]. Several clinical studies suggested that HNSCC patients with hypoxic tumors have a significant reduction in RT effectiveness [7–10]. On the other hand, accumulating new evidence showed that metabolic rewiring may also contribute to tumor radioresistance [11]. Several studies reported that increased glycolytic activities are associated with poor HNSCC RT responses [12–14]. Mitochondria can also affect HNSCC radiosensitivity by regulating the antioxidant activity and ROS production [15–17]. A recent in vitro cell study showed that radioresistant HNSCC cells (rSCC-61) exhibit increased glucose uptake and reduced oxidative phosphorylation (OXPHOS) compared to their parental radiosensitive HNSCC cells (SCC-61) [18]. Moreover, hypoxia is closely linked to metabolic reprogramming, forming a complex and multifactorial interaction that drives tumor radioresistance [19]. Our recent in vitro cell study demonstrated that both Hypoxia-Inducible Factor 1-alpha (HIF-1α) and metabolic reprogramming play significant roles in radioresistance development for HNSCC [20]. Therefore, a comprehensive understanding of tumor hypoxia and metabolic rewiring together is critical for the development of targeted therapeutic strategies to improve clinical outcomes.

Several clinically available imaging techniques have been explored to measure tumor hypoxia and metabolic changes under RT. For instance, Positron Emission Tomography (PET) is employed to assess glucose metabolism (using radiotracer 18F-FDG) [21,22] and hypoxia (using radiotracer 18F-FMISO) [23,24] to predict radiotherapy outcomes. Magnetic Resonance Imaging (MRI) is utilized to assess tumor morphology, vascular changes, and tissue perfusion in response to RT [25], while Magnetic Resonance Spectroscopy (MRS) is explored to capture the intracellular metabolite changes under treatment [26,27]. However, both PET and MRI require expensive equipment and heavily rely on expertise, which makes them impractical for frequent and periodical measurements. Alternatively, optical spectroscopy has emerged as a promising non-invasive and cost-effective tool for assessing tumor hypoxia and metabolism in small tumors for cancer research [28–30]. Several groups have reported the use of optical spectroscopy techniques for tumor RT response monitoring using small animal models. Hu et al. reported that diffuse reflectance spectroscopy can be used to monitor tumor oxygenation and perfusion changes under fractional RT in head and neck tumors [31]. Diaz et al. utilized diffuse reflectance spectroscopy to study radiation-induced alterations in vascular parameters for lung cancer xenografts [32]. Dadgar et al. used diffuse reflectance spectroscopy to investigate the RT-induced reoxygenation for head and neck tumors [33]. However, all previous optical spectroscopy based tumor RT studies primarily focused on the monitoring of vascular parameters alone, which may not offer a holistic understanding of the radiotherapeutic response in radioresistant tumors [32,33]. A systems-level measure of both vascular and metabolic alterations under radiation is critical to provide new insights into the biological pathways involved in tumor radioresistance.

We previously reported the potential of optical spectroscopy for characterizing over five key metabolic and vascular parameters in flank tumor models [29,30]. In this study, we further demonstrated that our multi-parametric functional optical spectroscopy can be used for systematic monitoring of metabolic and vascular changes in HNSCC tumors in vivo under radiation stress. By using our portable optical spectroscopy, we quantified several functional parameters, including oxygen saturation (StO2), hemoglobin levels, glucose uptake, and mitochondrial membrane potential (MMP) on small tumors to understand HNSCC tumor responses to radiation treatment. We observed significant vascular and metabolic alterations in radioresistant HNSCC tumors following radiation, but minimal changes in radiosensitive HNSCC tumors. This finding suggests that vascular and metabolic changes may serve as useful biomarkers for future radioresistance prediction and therapeutic guidance for HNSCC management. Our multi-parametric functional optical spectroscopy has the potential to provide valuable insight into the prognostic evaluation and therapeutic guidance.

2. Materials and methods

2.1. Tumor xenografts and fractional radiation therapy

All in vivo animal experiments described in this study were performed according to a protocol approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Kentucky. A total of 10 male or female athymic nude mice (nu/nu, Jackson Laboratory) with an age of ∼8 weeks were used in this study. All mice were housed in an animal housing facility with access to food and water and standard 12-hour light/dark cycles. Animals were assigned to (1) the radioresistant tumor group (rSCC-61, n = 5); (2) the radiosensitive tumor group (SCC-61, n = 5). The SCC-61 cell is a radiosensitive human HNSCC cell line derived from the base of the tongue, while the rSCC-61 cell is a radioresistant cell line generated from SCC-61 cells [18,34]. Both SCC-61 and rSCC-61 cells used in this study were cultured in the DMEM/F12 medium (Gibco) supplemented with 10% FBS (Gibco) and 1X Penicillin Streptomycin (Gibco) at 37°C and 5% CO2. Mice received a subcutaneous injection of rSCC-61 or SCC-61 cells (1 × 106-2 × 106 cells in 100 µL PBS (Gibco) with Matrigel (Corning)) under anesthesia with isoflurane (1-2% v/v) in room air.

On day 10 after the cell injection, all tumors were characterized using our quantitative optical spectroscopy platform [35]. The optical measurements on both SCC-61 and rSCC-61 tumors prior to any radiation treatment served as the baseline control for each of the tumor groups. Two days post the baseline optical measurements, tumors were subjected to radiation treatment with five doses of 2 Gy over 10 consecutive days (10 Gy in total) using an X-RAD 225XL biological cabinet (Precision X-Ray, North Branford, CT). Animals were placed in the center of a 10 × 10 cm X-ray radiation field. During radiation treatment, mice were kept under anesthesia using ketamine while the entire animal body was covered with lead blocks except the tumor. Mice were monitored daily during the entire course of RT. Once a total of 10 Gy RT was given to the tumors, optical measurements were performed again on the tumors under isoflurane anesthesia. The optical measurements on both SCC-61 and rSCC-61 tumors post the whole session of RT served as the RT treated group for each of the tumor groups. The detailed timeline for cell injections, optical measurements, and fractional RT is illustrated in Fig. 1. Tumor size was measured before each set of optical measurements. Tumor volume was calculated using the equation: volume = (Length x Width x Height)/2 [36].

Fig. 1.

Fig. 1.

Open in a new tab

Timeline for tumor cell injections, optical measurements and fractional RT in this study.

2.2. Quantitative multi-parametric optical spectroscopy

Our portable multi-parametric optical spectroscopy described previously [35] was used to perform optical measurements in a darkroom. Briefly, our optical spectroscopy was built using a high-power white LED (SOLIS-3C, Thorlabs), a compact spectrometer (FLAME-T-VIS-NIR, Ocean Optics), and a custom-designed fiber probe along with proper optical filters as illustrated in Fig. 2. The fiber probe has a total of 19 illumination fibers that are bundled in the center of the common end and a total of 19 collection fibers that are distributed radially around the illumination fibers (Fig. 2, bottom left). The diameter of each fiber is 200  μm, and the source-detector separations of the probe cover a range from 0.2 mm to 1.8 mm. The diffuse reflectance portion of the platform was used to perform diffuse reflectance measurements on small tumors thereby reporting tumor vascular parameters, while the fluorescence portion of the platform was used to measure the fluorescence of two metabolic probes including 2-(N-(7-Nitrobenz-2-oxa-1,3-diazol-4-yl)amino)-2-deoxyglucose (2-NBDG to report glucose uptake) and Tetramethylrhodamine Ethyl Ester (TMRE to report MMP) thereby reporting tumor metabolism. All mice were fasted for 6 hours for optical measurements. All mice received a tail-vein injection of TMRE (100 μL of 100 μM in PBS) first and then a tail-vein injection of 2-NBDG (100 μL of 6 mM 2-NBDG in PBS) with a 20-minute delay [30]. Optical measurements were obtained by placing the fiber probe gently on tumors when the animals were anesthetized with 1-2% v/v isoflurane. Prior to any probe injection, the baseline diffuse reflectance and fluorescence spectra were measured on tumors. For the baseline measurements, three to five locations (depending on tumor size) were measured and then averaged for data processing. For the longitudinal kinetic curve monitoring measurements, the fiber probe was fixed on the same tissue site (central region of the tumor), so one single point measure per time point was collected for a total of 80 minutes for each animal. All measurements on each animal were acquired in a darkroom. Diffuse reflectance spectra from 450 nm to 650 nm were acquired (with an integration time of 8 ms) to report tumor vascular parameters. Fluorescence spectra of 2-NBDG (from 520 nm to 600 nm, excited by 450 nm light) and TMRE (from 565 nm to 650 nm, excited by 550 nm light) were acquired with an integration time of 1 second to report tumor metabolism. Optical signals were averaged across all the detector fibers and then delivered to the spectrometer via a seven-round to linear fiber bundle (BFL105HS02, Thorlabs).

Fig. 2.

Fig. 2.

Open in a new tab

(A). Schematic of the optical spectroscopy setup. Bottom left: custom-designed fiber-optics probe; Bottom right: photo of a typical xenograft tumor under optical measurement. (B) photo of the actual spectroscopy system in a stand-alone portable cart.

2.3. Spectral data analysis and statistics

All raw diffuse reflectance spectra were calibrated using a 20% reflectance standard (Spectralon, Labsphere) before quantitative analysis. Our previously reported scalable inverse Monte Carlo (MC) model [35,37–39] was used to process the calibrated diffuse reflectance and fluorescence spectra, thereby extracting the tissue absorption, scattering, native fluorescence of 2-NBDG and TMRE from the optical spectra measured on small tumors. Briefly, each diffuse reflectance spectrum was processed by our MC inversion model to extract its corresponding absorption and scattering spectrum. During this process, the MC inversion model adaptively fits the measured reflectance spectrum to the MC simulated spectra (where both absorption and scattering properties were knowns) until the sum of squares error between the two is minimized, then the corresponding absorption coefficients spectrum and scattering coefficient spectrum were extracted for the specific diffuse reflectance spectrum. The extracted absorption spectra were further processed to extract tissue vascular parameters using a fitting procedure with a linear combination of the extinction spectra of oxy-hemoglobin and deoxy-hemoglobin based on Beer’s law [40]. Each raw fluorescence spectrum was processed by our MC inversion model to extract the intrinsic fluorescence spectrum. The MC model assumes that the measured raw fluorescence is a function of three parameters including fluorophore concentration, absorbed energy probability, and fluorescence escape probability [37]. The absorbed energy probability and the fluorescence escape probability are dependent on tissue optical properties at both excitation wavelength and emission wavelength [37], therefore, they can be estimated based on the formerly extracted absorption and scattering information. A two-step inversion was required for fluorescence data processing. The first step was to process the diffuse reflectance spectrum to extract absorption and scattering coefficient spectra, the second step was to utilize these extracted tissue optical properties with the fluorescence MC inversion model to extract the intrinsic fluorescence spectrum. The MC extracted intrinsic 2-NBDG and TMRE fluorescence spectra were used to estimate the glucose uptake and MMP. Specifically, the mean of the peak-band fluorescence intensity of intrinsic 2-NBDG (540 nm ± 5 nm) and TMRE (580 nm ± 5 nm) spectra were used to represent the 2-NBDG and TMRE signals. The 2-NBDG and TMRE signals taken at different time points were used to create the kinetic uptake curves. The 2-NBDG uptake at 60 minutes post 2-NBDG injection (2-NBDG60) and the TMRE uptake at 80 minutes post TMRE injection (TMRE80) that measured at the same time point were used to report final 2-NBDG and TMRE uptake. The characterization of our optical spectroscopy and MC model has been reported in detail recently [35]. The fluorescence intensities, hemoglobin contents, StO2, or average scattering among different groups were compared with an unpaired (for unpaired study) or paired (for paired study) two-sample t-test. A p-value < 0.05 was considered to be statistically different between the two groups under comparison. MATLAB (Mathworks, USA) was used to perform all data processing and statistical analysis.

3. Results

3.1. Baseline optical properties of in vivo small tumors

Figure 3(A) showed representative photos of a radioresistant xenograft tumor (rSCC-61) and a radiosensitive xenograft tumor (SCC-61), that were measured using our multi-parametric optical spectroscopy prior to any radiation treatment. Figure 3(B) showed tumor volume information, while Fig. 3(C) showed the average of calibrated diffuse reflectance spectra measured on the two types of xenograft tumors. Figure 3(B) showed that the tumor volumes of rSCC-61 tumors were larger than that of SCC-61 tumors, though they were all measured on day 10 post tumor cell injection. As shown in Fig. 3(C), we noticed slightly lower variations of the spectra intensities in the rSCC-61 tumors compared to the SCC-61 tumors. We also observed spectra shape differences around the 550 nm band between the two tumor groups, which were likely caused by the different oxygenation levels in the two types of tumors. Figure 3(D)-(E) showed the MC estimated optical properties of rSCC-61 tumors and SCC-61 tumors. Figure 3(D) showed the comparison of the averaged absorption coefficient spectra (480-620 nm band) between the two types of xenograft tumors. Figure 3(D) showed that rSCC-61 tumors had significantly lower absorption coefficients compared to SCC-61 tumors. Figure 3(E) showed the comparison of the average reduced scattering coefficient spectra (480-620 nm band) between the rSCC-61 tumors and the SCC-61 tumors. It appears SCC-61 tumors had higher reduced scattering levels (480-620 nm band) compared to rSCC-61 tumors.

Fig. 3.

Fig. 3.

Open in a new tab

(A) Representative SCC-61 and rSCC-61 tumors measured by our optical spectroscopy; (B) Tumor volumes of SCC-61 and rSCC-61 tumors on day 10 post cell injection; (C) Average diffuse reflectance spectra measured on all five in vivo SCC-61 tumors and five in vivo rSCC-61 tumors; (D) MC model extracted absorption coefficients of five in vivo SCC-61 tumors and five in vivo rSCC-61 tumors; (E) MC model extracted reduced scattering coefficients of five in vivo SCC-61 tumors and five in vivo rSCC-61 tumors. Tumor volume was estimated using the equation: volume = (Length x Width x Height)/2.

3.2. Baseline vascular and metabolic parameters of in vivo small tumors

Figure 4 showed the vascular and metabolic endpoints for both rSCC-61 and SCC-61 tumors prior to any radiation treatment. These vascular parameters were estimated from the MC extracted absorption spectra as shown in Fig. 3(D), while the metabolic parameters including glucose uptake (2-NBDG uptake) and MMP (TMRE uptake) were estimated from the MC corrected fluorescence spectra of 2-NBDG and TMRE measured on in vivo tumors. Figure 4(A) showed that rSCC-61 tumors had significantly lower StO2 compared to SCC-61 tumors, which was expected as rSCC-61 tumors had larger tumor volumes compared to SCC-61 tumors as shown in Fig. 3(B). Figure 4(B) showed that rSCC-16 tumors had significantly lower total hemoglobin concentration ([THB]) compared to SCC-61 tumors. Figure 4(C) showed that rSCC-61 tumors had comparable deoxygenated hemoglobin concentration ([HB]) compared to SCC-61 tumors, while Fig. 4(D) showed that rSCC-61 tumors had significantly lower oxygenated hemoglobin concentration ([HbO2]) compared to SCC-61 tumors.

Fig. 4.

Fig. 4.

Open in a new tab

MC model extracted tissue vascular parameters including (A) StO2, (B) [THB], (C) [HB], (D) [HbO2] for both SCC-61 and rSCC-61 tumors. (E) 2-NBDG uptake kinetics over 60 minutes and (F) boxplot of 2-NBDG uptake at 60 minutes (2-NBDG60). (G) TMRE uptake kinetics over 80 minutes and (H) boxplot of TMRE uptake at 80 minutes (TMRE80). A student t-test was used for all statistical analyses.

Figure 4(E) showed the fluorescence uptake kinetics of 2-NBDG over 60 minutes for the two types of tumors. Figure 4(E) showed that 2-NBDG fluorescence in SCC-61 tumors reached its peak at 2 minutes post injection, while 2-NBDG fluorescence in rSCC-61 tumors reached its peak at 4 minutes post injection. This interesting phenomenon may reflect the different hemodynamics in the two tumor types. The rSCC-61 tumors exhibited higher 2-NBDG uptake compared to the SCC-61 tumors over 60 minutes. The boxplot in Fig. 4(F) showed that the delivery corrected 2-NBDG60 uptake is significantly higher in rSCC-61 tumors compared to SCC-61 tumors (p < 0.002). Similarly, Fig. 4(G) showed the TMRE fluorescence uptake kinetics over 80 minutes. The intensity of TMRE fluorescence gradually increased and then stabilized in both tumor groups. The signal stabilized at approximately 50 minutes post injection for both groups, with the rSCC-61 tumors exhibiting a higher TMRE uptake compared to the SCC-61 tumors. The boxplot in Fig. 4(H) showed that the TMRE80 uptake was significantly lower in rSCC-61 tumors compared to SCC-61 tumors (p < 0.001).

3.3. Optical properties changes of radioresistant and radiosensitive tumors post RT

Figure 5(A)-(D) showed the changes in tumor volumes, the averaged calibrated diffuse reflectance spectra, and the corresponding optical properties of SCC-61 tumors, while Fig. 5(E)-(H) showed these changes of rSCC-61tumors prior and post a total of 10 Gy radiation treatment. Figure 5(A) showed that volumes of SCC-61 tumors increased slightly post radiation treatment but not statistically significant (p = 0.29), which suggested that the RT has slowed the tumor growth. In contrast, Fig. 5(E) showed that volumes of rSCC-61 tumors increased significantly (p < 0.02) post the radiation treatment, which suggested that the RT was not effective on the rSCC-61 tumors. Figure 5(B) showed that neither the average diffuse reflectance spectra intensity (480-620 nm) nor the spectra shape around the 550 nm band changed in SCC-61 tumors post radiation treatment. In contrast, Fig. 5(F) showed that both the average diffuse reflectance spectra intensity (480-620 nm) and the spectra shape around the 550 nm band changed significantly in rSCC-61 tumors post RT. Figure 5(C) showed that the SCC-61 tumors had significantly decreased absorption coefficients (480-620 nm) post the radiation treatment, while Fig. 5(G) showed that the rSCC-61 tumors had significantly increased absorption coefficients (480-620 nm) post the radiation treatment. Figure 5(D) and (H) showed that both SCC-61 tumors and rSCC-61 tumors had increased averaged reduced scattering coefficients (480-620 nm band) post radiation treatment, but it was more obvious for rSCC-61 tumors (Fig. 5(H)).

Fig. 5.

Fig. 5.

Open in a new tab

Tumor volumes of SCC-61 (A) and rSCC-61(E) tumors prior to and post RT. Average diffuse reflectance spectra measured on SCC-61 (B) and rSCC-61 tumors (F), the MC model extracted absorption coefficients of SCC-61 (C) and rSCC-61 (G) tumors and the MC model extracted reduced scattering coefficients of SCC-61 (D) and rSCC-61 (H) tumors prior to and post radiation treatment. Tumor volume was estimated using the equation: volume = (Length x Width x Height) / 2. One animal from the SCC-61 group showed obvious tumor necrosis, therefore, the animal was euthanized before RT treatment per IACUC protocol.

3.4. Radiation-induced vascular changes of radioresistant and radiosensitive tumors

Figure 6 showed the changes of vascular parameters for both SCC-61 tumors and rSCC-61 tumors post a total of 10 Gy radiation treatment. These vascular parameters were estimated from the MC extracted absorption spectra as shown in Fig. 5(C) and (G), respectively. Figure 6(A)-(D) showed that there were minimal changes in StO2, [THB], and [HbO2] for SCC-61 tumors post radiation treatment. The [HB] in SCC-61 tumors decreased slightly post radiation treatment, but not statistically significant (p = 0.07). In contrast, Fig. 6(E)-(H) showed that there were dramatic changes in StO2, [THB], and [HbO2] for rSCC-61 tumors post radiation treatment. Specifically, both the StO2 and [HbO2] in rSCC-61 tumors increased significantly post radiation treatment, with a p-value less than 0.002 and 0.02, respectively. The increase of StO2 suggested obvious reoxygenation in rSCC-61 tumors under the radiation stress, while the increase of [HbO2] suggested a significantly increased oxygen delivery rate under the radiation treatment. The [THB] in rSCC-61 tumors also increased post radiation treatment with a p-value close to statistically significant (p = 0.05), while the [HB] in rSCC-61 tumors did not change post radiation treatment (p = 0.25). The increase of [THB] suggested obviously increased blood contents in rSCC-61 tumors under radiation stress. In summary, there were dramatic vascular changes induced by radiation stress in rSCC-61 tumors but not in SCC-61 tumors. Specifically, rSCC-61 tumors showed increased StO2 (indicating re-oxygenation), [THB] (indicating blood perfusion), and [HbO2] (indicating oxygen supply) under the radiation stress.

Fig. 6.

Fig. 6.

Open in a new tab

MC models extracted tissue vascular parameters including (A and E) StO2, (B and F) [THB], (C and G) [HB], (D and H) [HbO2] for both SCC-61 and rSCC-61 tumors prior to and post RT. A paired Student’s t-test was used for all statistical analyses.

3.5. Radiation-induced metabolic changes of radioresistant and radiosensitive tumors

Figure 7 showed the changes of metabolic parameters for both SCC-61 tumors and rSCC-61 tumors post a total of 10 Gy radiation treatment. These metabolic parameters were estimated from the MC-corrected fluorescence spectra of 2-NBDG and TMRE measured on in vivo tumors. Figure 7(A) showed the fluorescence uptake kinetics curves of 2-NBDG over 60 minutes for the SCC-61 tumors prior to and post 10 Gy radiation treatment, while Fig. 7(B) showed the corresponding boxplot of the delivery-corrected 2-NBDG60 uptake. It appears that there were minimal changes in the kinetics and intensity of 2-NBDG60 uptake for SCC-61 tumors with the radiation treatment. In contrast, Figs. 7(E) and (F) showed significant changes in both the kinetics and intensity of 2-NBDG60 uptake for rSCC-61 tumors with the radiation treatment. The delivery-corrected 2-NBDG60 uptake was decreased post radiation treatment, with a p-value close to statistically significant (p = 0.05). Figures 7(C) and (G) showed the fluorescence uptake kinetics curves of TMRE over 80 minutes for the SCC-61 and rSCC-61 tumors prior and post 10 Gy radiation treatment, while Figs. 7(D) and (H) showed the corresponding boxplots of the TMRE80 uptake for the two types of tumors with radiation treatment. It appears that the TMRE80 uptake had a significant increase in rSCC-61 tumors (p < 0.05) but no changes in SCC-61 tumors post radiation. In summary, there were dramatic metabolic changes induced by radiation stress in rSCC-61 tumors but not in SCC-61 tumors. Specifically, rSCC-61 tumors showed decreased glucose uptake (2-NBDG uptake) and enhanced mitochondrial activities (TMRE uptake) post radiation treatment.

Fig. 7.

Fig. 7.

Open in a new tab

(A) 2-NBDG uptake kinetics over 60 minutes and (B) boxplot of 2-NBDG uptake at 60 minutes (2-NBDG60), and (C) TMRE uptake kinetics over 80 minutes and (D) boxplot of TMRE uptake at 80 minutes (TMRE80) in SCC-61 tumors. (E) 2-NBDG uptake kinetics over 60 minutes and (F) boxplot of 2-NBDG uptake at 60 minutes (2-NBDG60) and (G) TMRE uptake kinetics over 80 minutes and (H) boxplot of TMRE uptake at 80 minutes (TMRE80) in rSCC-61 tumor prior to and post RT. A paired Student’s t-test was used for all statistical analyses.

4. Discussion

Radioresistance remains a significant clinical challenge in HNSCC, leading to tumor recurrence and poor therapeutic outcomes [41]. Metabolic reprogramming and tumor hypoxia have been recognized as the hallmarks of cancer and may play critical roles in tumor radioresistance development. Understanding the roles of tumor hypoxia and metabolic reprogramming underlying radioresistance is critical for developing effective RT strategies and improving treatment efficacy [19]. Despite the wide exploration of various imaging techniques to measure hypoxia and metabolic alterations in both clinical and preclinical models, there remains a significant unmet need for a practical tool that can provide a thorough measurement of vascular and metabolic changes under radiation stress. In this study, we demonstrated a portable multi-parametric functional spectroscopy to quantify the key vascular and metabolic parameters in small animal models under radiation treatment. For the first time, we reported the vascular and metabolic changes within matched models of radiation resistance in HNSCC (SCC-61 and rSCC-61) post a total of 10 Gy fractional radiation. Our results showed dramatic vascular and metabolic changes in radioresistant HNSCC tumors prior to and post radiation treatment, but not in radiosensitive HNSCC tumors.

The baseline vascular features of the radiosensitive and radioresistant tumors in Fig. 4 revealed distinct differences in oxygenation and tumor perfusion. We observed that rSCC-61 tumors have significantly lower StO2, [THB], and [HbO2] in comparison to SCC-61 tumors. This suggested that rSCC-61 tumors had a more hypoxic microenvironment with reduced oxygen delivery capacity, which could be attributed to larger volumes of rSCC-61 tumors. The rSCC-61 group grew larger tumors at day 10 post cell injection, which is most likely due to the higher proliferation than the SCC-61 group as reported previously [18]. The metabolic baseline results also revealed distinct metabolic demand in rSCC-61 tumors compared to SCC-61 tumors. We observed that rSCC-61 had significantly higher glucose uptake and MMP than SCC-61 tumors, which agrees with the results in our previous microscopy study [20] and published the Seahorse Assay study [18]. Sun et al. also suggested that radioresistant tumors have lower oxidative metabolism and higher glucose uptake than radiosensitive tumors in HNSCC patients [42].

To explore the tumor vascular responses under radiation stress, a total of 10 Gy fractional radiation was utilized to treat both SCC-61 and rSCC-61 tumors. The tumor volume data in Fig. 5 revealed a significant increase in tumor size in rSCC-61 tumors while a minimal increase in SCC-61 tumors, suggesting an effective growth delay in SCC-61 tumors but ineffective treatment in rSCC-61 tumors using fractional RT. We observed notable variations in vascular changes between rSCC-61 and SCC-61 tumors in response to fractional RT as shown in Fig. 6. Specifically, we observed a significantly increased StO2 post RT in rSCC-61 tumors which aligned well with previous similar studies [31–33], suggesting that reoxygenation post RT can be one of the effective biomarkers for identifying radioresistant tumors [43]. Meanwhile, we captured an increased [THB] in rSCC-61 tumors, which agreed with several former studies reporting increased tumor perfusion following radiation [44–47]. We also observed that rSCC-61 tumors had increased [HbO2], which is consistent with other studies showing an increased [HbO2] in radioresistant tumors under RT [32]. In contrast, SCC-61 tumors have minimal changes in vascular parameters except for a slight reduction in [HB]. Reduced [HB] reflected decreased oxygen consumption in local tumor tissues thereby indicating an increased tumor death [31,48], which aligned well with the delayed growth in SCC-61 tumors post RT.

In addition to vascular changes, we were able to recapitulate the metabolic alterations between SCC-61 tumors and rSCC-61 tumors in response to fractional RT. We noted that SCC-61 tumors had a minimal change in either glucose uptake or MMP prior to and post RT. However, rSCC-61 tumors had significantly reduced glucose uptake and enhanced MMP post radiation treatment, suggesting a significant metabolic shift from glycolysis to OXPHOS. Our in vivo animal data is different compared with the metabolic trend found in our previous in vitro study where we see significant metabolic changes for both rSCC-61 and SCC-61 cells [20]. This inconsistent trend is likely due to the inherent difference in the tumor microenvironment (in vitro cells and in vivo tumors). However, a former clinical study observed that radioresistant HNSCC tumors have increased oxidative metabolism under radiation [42]. This interesting study may support the metabolic trend we observed here, where the rSCC-61 tumors shift the glycolysis to OXPHOS under radiation stress. Overall, the distinct metabolic responses observed in radioresistant tumors post radiation indicated that mitochondrial function may potentially be an effective therapeutic target for radio-sensitizing and improving RT efficacy.

By combining diffuse reflectance and fluorescence measurements of 2-NBDG and TMRE, our approach enables longitudinal assessment of tissue optical properties, vascular parameters, and metabolic endpoints in a cost-effective and easy-to-access way. Our techniques can provide a comprehensive and quantitative evaluation of tumor functional changes in response to radiation and bridge the gap between vascular and metabolic alterations to radiation resistance. We will utilize our techniques combined with metabolic inhibitors involved in glycolysis and mitochondrial metabolic pathways to explore the potentially effective radiotherapeutic target as a radiosensitizer for managing radioresistant tumors in our future therapeutic study. Additionally, reoxygenation and mitochondrial activity exhibited a good prognostic relevance in tumor radiotherapy outcomes, which can be further explored as effective indicators for predicting radioresistant and clinical outcomes in our future diagnostic study. Our former study has demonstrated that the particular fiber-probe geometry used in this study can report the metabolic and vascular differences between the two types of HNSCC tumors in flank models [35], therefore we directly utilized this well-characterized fiber probe geometry for this present study due to the practical convenience. It should be noted that our fiber probe has a sensing range from the superficial skin layer to the shallow region of the small tumors. However, a fiber-probe with a better sensing depth should be considered in future study to probe the deep region of tumor tissues. Our portable multi-parametric functional optical spectroscopy may offer a cost-effective, and easy-to-access in characterizing tumor physiology and monitoring treatment response, which could be used for radioresistant evaluation and therapeutic guidelines in HNSCC clinical translation study.

5. Conclusion

We demonstrated a multi-parametric functional optical spectroscopy to monitor the vascular and metabolic changes of radiosensitive and radioresistant HNSCC tumors with fractional RT. We found that radioresistant and radiosensitive HNSCC tumors had significantly different vascular and metabolic changes in response to fractional radiation stress. Specifically, radioresistant HNSCC tumors had significant reoxygenation and obvious metabolic shift from glycolysis to OXPHOS under RT. In contrast, radiosensitive HNSCC tumors had minimal changes in either vascular or metabolic changes under RT. Looking at both reoxygenation and mitochondrial function may provide greater insight on the prediction of radioresistant tumors, and the mitochondrial metabolism pathway may potentially be an effective therapeutic target to improve RT outcomes. Our multi-parametric functional spectroscopy could be a useful technique for longitudinally studying the roles of tumor vascular microenvironment and metabolic rewiring in tumor radioresistance in a more cost-effective and non-invasive manner.

Acknowledgments

This work was supported by NIDCR/NIGMS-R01DE031998. The funders had no role in study design, data collection, analysis, decision to publish, or preparation of the manuscript.

Funding

National Institute of Dental and Craniofacial Research 10.13039/100000072 ( R01DE031998); National Institute of General Medical Sciences 10.13039/100000057 ( R01DE031998).

Disclosures

The authors declare that there are no known conflicts of interest related to this article.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

  • 1. Barsouk A., Aluru J. S., Rawla P., et al. , “Epidemiology, risk factors, and prevention of head and neck squamous cell carcinoma,” Med. Sci. 11(2), 42 (2023). 10.3390/medsci11020042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.“Cancer facts & figures 2024,” (2024).
  • 3. Marur S., Forastiere A. A., “Head and neck cancer: Changing epidemiology, diagnosis, and treatment,” Mayo Clin. Proc. 83(4), 489–501 (2008). 10.4065/83.4.489 [DOI] [PubMed] [Google Scholar]
  • 4. Bouleftour W., Rowinski E., Louati S., et al. , “A review of the role of hypoxia in radioresistance in cancer therapy,” Med. Sci. Monit. 27, e934116 (2021). 10.12659/MSM.934116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Wang H., Jiang H., Van De Gucht M., et al. , “Hypoxic Radioresistance: Can ROS Be the Key to Overcome It?” Cancers 11(1), 112 (2019). 10.3390/cancers11010112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Grimes D. R., Partridge M., “A mechanistic investigation of the oxygen fixation hypothesis and oxygen enhancement ratio,” Biomed. Phys. Eng. Express 1(4), 045209 (2015). 10.1088/2057-1976/1/4/045209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Nordsmark M., Overgaard M., Overgaard J., “Pretreatment oxygenation predicts radiation response in advanced squamous cell carcinoma of the head and neck,” Radiother. Oncol. 41(1), 31–39 (1996). 10.1016/S0167-8140(96)91811-3 [DOI] [PubMed] [Google Scholar]
  • 8. Brizel D. M., Sibley G. S., Prosnitz L. R., et al. , “Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck,” Int. J. Radiat. Oncol. Biol. Phys. 38(2), 285–289 (1997). 10.1016/S0360-3016(97)00101-6 [DOI] [PubMed] [Google Scholar]
  • 9. Nordsmark M., Overgaard J., “A confirmatory prognostic study on oxygenation status and loco-regional control in advanced head and neck squamous cell carcinoma treated by radiation therapy,” Radiother. Oncol. 57(1), 39–43 (2000). 10.1016/S0167-8140(00)00223-1 [DOI] [PubMed] [Google Scholar]
  • 10. Nordsmark M., Bentzen S. M., Rudat V., et al. , “Prognostic value of tumor oxygenation in 397 head and neck tumors after primary radiation therapy. An international multi-center study,” Radiother. Oncol. 77(1), 18–24 (2005). 10.1016/j.radonc.2005.06.038 [DOI] [PubMed] [Google Scholar]
  • 11. Yu Y., Yu J., Ge S., et al. , “Novel insight into metabolic reprogrammming in cancer radioresistance: A promising therapeutic target in radiotherapy,” Int. J. Biol. Sci. 19(3), 811–828 (2023). 10.7150/ijbs.79928 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Jung Y. S., Najy A. J., Huang W., et al. , “HPV-associated differential regulation of tumor metabolism in oropharyngeal head and neck cancer,” Oncotarget 8(31), 51530–51541 (2017). 10.18632/oncotarget.17887 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Yan S. X., Luo X. M., Zhou S. H., et al. , “Effect of antisense oligodeoxynucleotides glucose transporter-1 on enhancement of radiosensitivity of laryngeal carcinoma,” Int. J. Med. Sci. 10(10), 1375–1386 (2013). 10.7150/ijms.6855 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Kunkel M., Moergel M., Stockinger M., et al. , “Overexpression of GLUT-1 is associated with resistance to radiotherapy and adverse prognosis in squamous cell carcinoma of the oral cavity,” Oral Oncol. 43(8), 796–803 (2007). 10.1016/j.oraloncology.2006.10.009 [DOI] [PubMed] [Google Scholar]
  • 15. Grasso D., Medeiros H. C. D., Zampieri L. X., et al. , “Fitter Mitochondria Are Associated With Radioresistance in Human Head and Neck SQD9 Cancer Cells,” Front. Pharmacol. 11, 263 (2020). 10.3389/fphar.2020.00263 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Qu Y., Zhang H., Zhao S., et al. , “The effect on radioresistance of manganese superoxide dismutase in nasopharyngeal carcinoma,” Oncol. Rep. 23(4), 1005–1011 (2010). 10.3892/or_00000726 [DOI] [PubMed] [Google Scholar]
  • 17. Zhu C. G., Martinez A. F., Martin H. L., et al. , “Near-simultaneous intravital microscopy of glucose uptake and mitochondrial membrane potential, key endpoints that reflect major metabolic axes in cancer,” Sci. Rep. 7(1), 13772 (2017). 10.1038/s41598-017-14226-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Mims J., Bansal N., Bharadwaj M. S., et al. , “Energy metabolism in a matched model of radiation resistance for head and neck squamous cell cancer,” Radiat. Res. 183(3), 291–304 (2015). 10.1667/RR13828.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Shi Z., Hu C., Zheng X., et al. , “Feedback loop between hypoxia and energy metabolic reprogramming aggravates the radioresistance of cancer cells,” Exp. Hematol. Oncol. 13(1), 55 (2024). 10.1186/s40164-024-00519-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Yan J., Goncalves C. F. L., Saha P. S., et al. , “Optical imaging provides flow-cytometry-like single-cell level analysis of HIF-1α-mediated metabolic changes in radioresistant head and neck squamous carcinoma cells,” Biophotonics Discov. 2(01), 012702 (2025). 10.1117/1.BIOS.2.1.012702 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Wilson G. D., Thibodeau B. J., Fortier L. E., et al. , “Glucose metabolism gene expression patterns and tumor uptake of 18F-fluorodeoxyglucose after radiation treatment,” Int. J. Radiat. Oncol. Biol. Phys. 90(3), 620–627 (2014). 10.1016/j.ijrobp.2014.06.062 [DOI] [PubMed] [Google Scholar]
  • 22. Roach M. C., Turkington T. G., Higgins K. A., et al. , “FDG-PET assessment of the effect of head and neck radiotherapy on parotid gland glucose metabolism,” Int. J. Radiat. Oncol. Biol. Phys. 82(1), 321–326 (2012). 10.1016/j.ijrobp.2010.08.055 [DOI] [PubMed] [Google Scholar]
  • 23. Rajendran J. G., Mankoff D. A., O’Sullivan F., et al. , “Hypoxia and glucose metabolism in malignant tumors: evaluation by [18F]fluoromisonidazole and [18F]fluorodeoxyglucose positron emission tomography imaging,” Clin. Cancer Res. 10(7), 2245–2252 (2004). 10.1158/1078-0432.CCR-0688-3 [DOI] [PubMed] [Google Scholar]
  • 24. Carles M., Fechter T., Grosu A. L., et al. , “(18)F-FMISO-PET Hypoxia Monitoring for Head-and-Neck Cancer Patients: Radiomics Analyses Predict the Outcome of Chemo-Radiotherapy,” Cancers 13(14), 3449 (2021). 10.3390/cancers13143449 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. van der Hulst H. J., Jansen R. W., Vens C., et al. , “The prediction of biological features using magnetic resonance imaging in head and neck squamous cell carcinoma: a systematic review and meta-analysis,” Cancers 15(20), 5077 (2023). 10.3390/cancers15205077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Chia M. L., Bulat F., Gaunt A., et al. , “Metabolic imaging distinguishes ovarian cancer subtypes and detects their early and variable responses to treatment,” Oncogene 44(9), 563–574 (2025). 10.1038/s41388-024-03231-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Chawla S., Kim S., Loevner L. A., et al. , “Proton and phosphorous MR spectroscopy in squamous cell carcinomas of the head and neck,” Acad. Radiol. 16(11), 1366–1372 (2009). 10.1016/j.acra.2009.06.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Rajaram N., Reesor A. F., Mulvey C. S., et al. , “Non-invasive, simultaneous quantification of vascular oxygenation and glucose uptake in tissue,” PLoS One 10(1), e0117132 (2015). 10.1371/journal.pone.0117132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Zhu C., Martin H. L., Crouch B. T., et al. , “Near-simultaneous quantification of glucose uptake, mitochondrial membrane potential, and vascular parameters in murine flank tumors using quantitative diffuse reflectance and fluorescence spectroscopy,” Biomed. Opt. Express 9(7), 3399–3412 (2018). 10.1364/BOE.9.003399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Zhu C., Li M., Vincent T., et al. , “Simultaneous in vivo optical quantification of key metabolic and vascular endpoints reveals tumor metabolic diversity in murine breast tumor models,” J. Biophotonics 12(4), e201800372 (2019). 10.1002/jbio.201800372 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Hu F., Vishwanath K., Salama J. K., et al. , “Oxygen and Perfusion Kinetics in Response to Fractionated Radiation Therapy in FaDu Head and Neck Cancer Xenografts Are Related to Treatment Outcome,” Int. J. Radiat. Oncol. Biol. Phys. 96(2), 462–469 (2016). 10.1016/j.ijrobp.2016.06.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Diaz P. M., Jenkins S. V., Alhallak K., et al. , “Quantitative diffuse reflectance spectroscopy of short-term changes in tumor oxygenation after radiation in a matched model of radiation resistance,” Biomed. Opt. Express 9(8), 3794–3804 (2018). 10.1364/BOE.9.003794 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Dadgar S., Troncoso J. R., Siegel E. R., et al. , “Spectroscopic investigation of radiation-induced reoxygenation in radiation-resistant tumors,” Neoplasia 23(1), 49–57 (2021). 10.1016/j.neo.2020.11.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Bansal N., Mims J., Kuremsky J. G., et al. , “Broad Phenotypic Changes Associated with Gain of Radiation Resistance in Head and Neck Squamous Cell Cancer,” Antioxid. Redox Signaling 21(2), 221–236 (2014). 10.1089/ars.2013.5690 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Hasan M. Z., Yan J., Zhu C., “Point-of-care optical spectroscopy platform and ratio-metric algorithms for rapid and systematic functional characterization of biological models in vivo,” J. Biomed. Opt. 29(12), 125002 (2024). 10.1117/1.JBO.29.12.125002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Tomayko M. M., Reynolds C. P., “Determination of subcutaneous tumor size in athymic (nude) mice,” Cancer Chemother. Pharmacol. 24(3), 148–154 (1989). 10.1007/BF00300234 [DOI] [PubMed] [Google Scholar]
  • 37. Palmer G. M., Ramanujam N., “Monte-Carlo-based model for the extraction of intrinsic fluorescence from turbid media,” J. Biomed. Opt. 13(2), 024017 (2008). 10.1117/1.2907161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Liu C. B., Rajaram N., Vishwanath K., et al. , “Experimental validation of an inverse fluorescence Monte Carlo model to extract concentrations of metabolically relevant fluorophores from turbid phantoms and a murine tumor model,” J. Biomed. Opt. 17(7), 0780031 (2012). 10.1117/1.JBO.17.7.077012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Saha P. S., Yan J., Zhu C., “Diffuse reflectance spectroscopy for optical characterizations of orthotopic head and neck cancer models in vivo,” Biomed. Opt. Express 15(7), 4176–4189 (2024). 10.1364/BOE.528608 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Franceschini M. A., Gratton E., Fantini S., “Noninvasive optical method of measuring tissue and arterial saturation: an application to absolute pulse oximetry of the brain,” Opt. Lett. 24(12), 829–831 (1999). 10.1364/OL.24.000829 [DOI] [PubMed] [Google Scholar]
  • 41. Alsahafi E., Begg K., Amelio I., et al. , “Clinical update on head and neck cancer: molecular biology and ongoing challenges,” Cell Death Dis. 10(8), 540 (2019). 10.1038/s41419-019-1769-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Sun A., Johansson S., Turesson I., et al. , “Imaging tumor perfusion and oxidative metabolism in patients with head-and-neck cancer using 1- [11C]-acetate PET during radiotherapy: preliminary results,” Int. J. Radiat. Oncol. Biol. Phys. 82(2), 554–560 (2012). 10.1016/j.ijrobp.2010.11.007 [DOI] [PubMed] [Google Scholar]
  • 43. Dietz A., Vanselow B., Rudat V., et al. , “Prognostic impact of reoxygenation in advanced cancer of the head and neck during the initial course of chemoradiation or radiotherapy alone,” Head Neck 25(1), 50–58 (2003). 10.1002/hed.10177 [DOI] [PubMed] [Google Scholar]
  • 44. Sunar U., Quon H., Durduran T., et al. , “Noninvasive diffuse optical measurement of blood flow and blood oxygenation for monitoring radiation therapy in patients with head and neck tumors: a pilot study,” J. Biomed. Opt. 11(6), 064021 (2006). 10.1117/1.2397548 [DOI] [PubMed] [Google Scholar]
  • 45. Goda F., Bacic G., O’Hara J. A., et al. , “The relationship between partial pressure of oxygen and perfusion in two murine tumors after X-ray irradiation: a combined gadopentetate dimeglumine dynamic magnetic resonance imaging and in vivo electron paramagnetic resonance oximetry study,” Cancer Res. 56(14), 3344–3349 (1996). [PubMed] [Google Scholar]
  • 46. Dewhirst M. W., Oliver R., Tso C. Y., et al. , “Heterogeneity in tumor microvascular response to radiation,” Int. J. Radiat. Oncol. Biol. Phys. 18(3), 559–568 (1990). 10.1016/0360-3016(90)90061-N [DOI] [PubMed] [Google Scholar]
  • 47. Sonveaux P., Dessy C., Brouet A., et al. , “Modulation of the tumor vasculature functionality by ionizing radiation accounts for tumor radiosensitization and promotes gene delivery,” FASEB j. 16(14), 1979–1981 (2002). 10.1096/fj.02-0487fje [DOI] [PubMed] [Google Scholar]
  • 48. Lee C. T., Boss M. K., Dewhirst M. W., “Imaging tumor hypoxia to advance radiation oncology,” Antioxid. Redox Signaling 21(2), 313–337 (2014). 10.1089/ars.2013.5759 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Articles from Biomedical Optics Express are provided here courtesy of Optica Publishing Group

RESOURCES