Open Access
Add to BibSonomy
Abstract
Immune checkpoint inhibitors have revolutionized cancer treatment. However, there are currently no methods for noninvasively and nondestructively evaluating tumor response to immune checkpoint inhibitors. We used diffuse reflectance spectroscopy to monitor in vivo tumor microenvironmental changes in response to immune checkpoint inhibitors in a CT26 murine colorectal cancer model. Mice growing CT26 tumor xenografts were treated with either anti-PD-L1, anti-CTLA-4, a combination of both inhibitors, or isotype control on 3 separate days. Monotherapy with either anti-PD-L1 or anti-CTLA-4 led to a large increase in tumor vascular oxygenation within the first 6 days. Reoxygenation in anti-CTLA-4-treated tumors was due to a combination of increased oxygenated hemoglobin and decreased deoxygenated hemoglobin, pointing to a possible change in tumor oxygen consumption following treatment. Within the anti-PD-L1-treated tumors, reoxygenation was primarily due to an increase in oxygenated hemoglobin with the minimal change in deoxygenated hemoglobin, indicative of a likely increase in tumor perfusion. The tumors in the combined treatment group did not show any significant changes in tumor oxygenation following therapy. These studies demonstrate the sensitivity of diffuse reflectance spectroscopy to tumor microenvironmental changes following immunotherapy and the potential of such non-invasive techniques to determine early tumor response to immune checkpoint inhibitors.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Cancer immunotherapy based on targeting immune checkpoint inhibitors, such as Cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and programmed cell death-1 (PD-1), has produced durable responses in some patients with colorectal cancer [1, 2]. While monotherapy consisting of either anti-CTLA-4 or anti-PD-1 is reasonably well-tolerated by patients, response rates are low with only ∼13 - 40% of patients showing objective response [3]. Combination therapy with both anti-CTLA-4 and anti-PD-1 leads to greater objective response in patients with ∼55% of patients showing complete response [4]. However, nearly the same percentage of patients show severe immune-related adverse events or toxicities, such as dermatitis, colitis, and hepatitis. Additionally, nearly 25% of initially responding patients develop secondary resistance to treatment. To date, there are few biomarkers that can accurately predict treatment response. Positive PD-L1 expression in tumor cells has been shown to be associated with increased objective response to anti-PD-1 therapy; however, the positivity threshold of PD-L1 expression to select patients for therapy varies across clinical centers and 10-20% of patients with negative PD-L1 expression have also seen clinical benefit [3]. The presence of tumor-infiltrating lymphocytes is associated with improved survival in colorectal cancer [5]. Mismatch repair-proficient colorectal cancer has been shown to be almost completely resistant to ICIs [2]. Immune profiling of biopsies acquired at several time points from accessible and progressing melanomas was found to be predictive of response to ICIs in melanoma patients [6]. However, all these techniques involve assessments of tumor biopsies at the cellular, gene, or protein level using destructive techniques. There are currently no methods that can noninvasively and nondestructively evaluate tumor response to ICIs. Personalized information about the functional and molecular state of the tumor can potentially determine if a tumor will be responsive to ICIs or is likely to develop secondary resistance after initial tumor regression.
Diffuse reflectance spectroscopy provides the ability to noninvasively interrogate and quantify longitudinal biological changes within tissue. DRS uses optical fibers to illuminate tissue with visible to near-infrared light and collect diffusely reflected light that has undergone multiple scattering and absorption events in tissue. The major contributors to tissue scattering (cells, nuclei, mitochondria, collagen) and absorption (oxygenated and deoxygenated hemoglobin, beta-carotene, melanin) are known to change with disease progression and are key contributors to our evaluation of disease pathology [7–9]. Based on the differential absorption properties of oxygenated (HbO2) and deoxygenated hemoglobin (dHb), DRS can measure the volume-averaged vascular oxygenation (sO2) within sampled tissue. We have demonstrated that DRS-based measurements of sO2 are inversely correlated with tumor hypoxia [10]. Simultaneous and real-time tracking of sO2 using DRS and tumor pO2 using oxygen-sensing microelectrodes in response to varying oxygen levels in inhaled air has shown strong concordance between both measurements [11]. Leveraging this sensitivity of DRS to changes in oxygenation, we and others have identified distinct longitudinal changes in tumor vascular oxygenation in response to radiation therapy in pre-clinical animal models [12–15] and patients [16–18] that could potentially help discriminate between treatment responders and non-responders. However, there are no studies using optical spectroscopy and imaging that have monitored tumor microenvironmental changes in response to immune checkpoint inhibitors.
The overall goal of this study was to determine longitudinal changes in the tumor microenvironment in response to immune checkpoint inhibitors. Tumors were treated with either control IgG antibodies, anti-CTLA-4, anti-PD-L1, or a combination of anti-CTLA-4 and anti-PD-L1. The choice of cell line here was driven by the early pre-clinical studies that utilized the CT26 cell line to investigate response to ICIs [19, 20]. We acquired optical spectra from the tumors every day for a period of 10 days and quantified microenvironmental parameters such as total hemoglobin concentration, vascular oxygenation, and tissue scattering. Our results demonstrate that ICIs cause significant changes in tumor vascular oxygenation and tissue scattering, and that the dynamics and source of these changes is dependent on the type of ICI therapy. Our studies show that DRS can potentially provide a noninvasive and nondestructive method to monitor functional changes with the tumor microenvironment in response to ICIs and identify responders and non-responders to treatment.
2. Materials and methods
2.1 Cell culture and tumor xenografts
The CT26 colorectal carcinoma cells were purchased from the American Type Culture Collection (Product: ATCC CRL-2638, Lot Number: 58494154) and passaged according to established protocols. Cells from third and fourth passages were used for tumor inoculation. Cells were injected into the hind flanks of 6-8-weeks-old Balb/c mice (n = 33) purchased from Jackson Laboratories. The mice were maintained at the University of Arkansas Central Laboratory Animal Facility with 12 h light/dark cycles and provided ad libitum access to food and water. All experiments were approved by the Institutional Animal Care and Use Committee at the University of Arkansas (IACUC protocols 18066, 20025). Mice were randomly assigned to one of four treatment groups - isotype (IgG control, n = 10), anti-mouse CTLA-4 (αCTLA4, n = 8), anti-mouse PD-L1 (αPDL1, n = 7), or a combination of αCTLA4 and αPDL1 (COMB, n = 8). A schematic of the study design is presented in Fig. 1(a).
Fig. 1. A. Study design depicting days of ICI treatment, optical measurements, and tumor excision. The red circles indicate treatment injection for the different groups. DRS was performed for 10 consecutive days. On days of treatment, DRS spectra were collected prior to ICI injection. B. Comparison of the fold change of tumor growth for the treatment groups - IgG, αCTLA-4, αPD-L1 and COMB. (C-F) Baseline, pre-treatment comparison of tumor volume, vascular oxygenation, total hemoglobin content and tissue scattering, respectively.
2.2 Immunotherapy agents
The immunotherapy agents were purchased from BioXcell (IgG: MPC-11, αCTLA4: UC10-4F10-11, αPDL1: 10F.9G2). Once tumor volumes reached between 80-120 mm3, mice were treated with 3 doses of 100 µg αCTLA-4, 200 µg αPD-L1, a combination of the two drugs, or 100 µg of IgG isotype control dissolved in 100 µL of sterile saline via intraperitoneal injection [19–21]. These doses were delivered on Days 1, 4, and 7 with Day 1 indicating the first day of treatment (Fig. 1(a)). Animals were euthanized 3 days after their last dosage and tumors were excised and snap-frozen for histopathology.
2.3 Diffuse reflectance spectroscopy and quantification of physiological parameters
The portable DRS instrument used in this study has been described in detail previously [10]. Briefly, it consists of a tungsten halogen lamp (HL-2000, Ocean Optics; Dunedin, FL) for illumination, a fiber optic spectrometer for spectral acquisition (Flame, Ocean Optics; Dunedin, FL), and a fiber optic probe (Fiber Tech Optica, Ontario, Canada). The probe consists of 4 source fibers (dia. = 200 μm, NA = 0.22) and five detector fibers located at 2.25 mm from the center of the source fibers. About 10-15 spectra in the wavelength range of 475-650 nm were acquired from each tumor on each day. DRS spectra were acquired every day for ten consecutive days. On days of treatment, spectra were acquired prior to ICI injection. We used an empirical lookup table (LUT)-based model [22, 23] developed specifically for this source-detector separation [10] to fit the acquired spectral data and quantify total hemoglobin concentration [Hb], oxygenated (HbO2) and deoxygenated hemoglobin (dHb), vascular oxygenation (sO2), and tissue scattering. Light scattering in tissue was assumed to have a negative power-law dependence on wavelength: ${{\mathrm \mu} }_{\rm s}^{{\prime}} \left( {\rm \lambda } \right) = {{\mathrm \mu} }_{\rm s}^{{\prime}} \left( {{\rm \lambda }_0} \right).\left( {{{\mathrm \lambda } \over {{\mathrm \lambda }_0}}} \right)^{-{\textrm B}}$, where λ0 is a reference wavelength at which light absorption is minimum and is set to 600 nm and µs’ is the reduced scattering coefficient. The absorption coefficient is calculated as the linear sum of absorption coefficients of individual absorbers, namely HbO2 and dHb, and animal skin: ${{\mathrm \mu} }_{\rm a}\left( {\mathrm \lambda } \right) = \left[ {{\rm Hb}} \right][{\rm \alpha }{\rm \sigma }_{{\rm Hb}{\rm O}_2}\left( {\mathrm \lambda } \right) + \left( {1-{ \; \alpha }){\sigma }_{{\rm dHb}}\left( {\mathrm \lambda } \right)} \right] + \left[ {{\rm Ml}} \right]{\rm mel}\left( {\mathrm \lambda } \right)$, where [Hb] and [Ml] respectively are total hemoglobin concentration and skin absorption. α is vascular oxygen saturation representing the ratio of oxygenated (HbO2) to total hemoglobin concentration [Hb]. The extinction coefficients of these absorbers have previously been established. The calculation of the absorption coefficient does include a correction factor for pigment packaging because we have shown such effects to be minimal at wavelengths above 500 nm [24]. Data analysis was performed in MATLAB (MathWorks, Natick, MA).
2.4 Statistical analysis
Optical properties are presented in this manuscript as either absolute values or fold changes. Raw optical properties from Day 1 through 10 were normalized by treatment group to the group mean of the baseline (Day 1 value). Fold change in each optical property is calculated by dividing the optical property on a given day by the pre-treatment baseline value on Day 1 (Day X/Day 1). A nested, two-way analysis of variance (ANOVA) was implemented to determine statistically significant differences in tissue scattering, hemoglobin concentration and vascular oxygenation across the four treatment groups. Post-hoc Tukey’s tests were used to determine statistical significance between specific groups.
2.5 Immunohistochemistry
Snap frozen tumors were sectioned on to glass slides and counterstained in Hematoxylin, dipped in differentiation solution, and blue buffer. The finished slides were mounted with Fluoromount G before imaging. A subset of snap frozen tumors was shipped to Johns Hopkins University where they were fixed in formalin, embedded in paraffin, and sectioned on to glass slides for histopathology and Masson’s trichrome staining for visualizing collagen. All slides were imaged using a Nikon fluorescence microscope using a 20X objective.
3. Results
The tumor growth curves for each treatment group are shown in Fig. 1(b). While there are no significant differences between groups over the 10-day period, it is evident that treatment with a combination of two ICIs (COMB) led to a slower growth rate in mice compared with monotherapy (αCTLA-4 or αPD-L1) or IgG control. There were no significant differences in the mean tumor volume at which treatment was initiated between the different groups (Fig. 1(c)). In addition, we found no significant differences in the optical properties of the different groups.
Figure 2 presents the fold-changes in vascular oxygenation (f-sO2; top row) and total hemoglobin concentration (bottom row) over the 10-day period for the three treatment groups – αPD-L1, αCTLA-4, and COMB. Data from the control IgG group is presented in each plot for one-to-one comparison with each treatment group. The mean sO2 in the IgG group decreased gradually over the 10-day period, reaching 0.5x of the Day 1 baseline at Day 10, with a concomitant increase in total hemoglobin concentration. Treatment with αPD-L1 or αCTLA-4 alone led to an increase in vascular oxygenation in the first 6 days. The mean f-sO2 in the αCTLA-4 group showed a large increase (∼1.5X baseline sO2) 24 hours after the first dose. The fold-change in sO2 was highest on Day 5, 24 hours after the second dose, and decreased thereafter to pre-treatment sO2 levels by Day 10. However, there were no significant differences between the IgG and αCTLA-4 groups on any day. Additionally, there was no significant change in cHb relative to baseline. We found a gradual increase in the mean f-sO2 in the αPD-L1 group, with a peak value of fold-change at Day 6 and a significant difference between the PD-L1 group and the IgG control group (p< 0.001). The fold-change in cHb on Day 10 was significantly higher compared with fold-change on the 5 previous days. Compared with the αPD-L1 and αCTLA-4 groups, there was minimal variation in mean sO2 and cHb in the COMB group over time.
Fig. 2. (A) Linear plots for percent change in vascular oxygenation (sO2) for αCTLA-4, αPD-L1 and COMB groups against the control group IgG across the 10 days of DRS. (B) Percent change in total hemoglobin content (Hb) for the four treatment groups across 10 days of spectroscopy. Data are presented as group mean (line) ±SEM (represented by error bars). Significant differences among treatments in specific days are illustrated with black pounds (#) while significant differences of specific days within groups are represented by asterisk of their respective color indicating a statistical significance at p < 0.05.
To investigate the source of changes in sO2 and cHb in each of the treatment groups, we analyzed the trends in HbO2 and dHb. The steady decrease observed in the IgG group can be attributed almost entirely to a large increase in dHb over time, suggesting an increase in oxygen demand and consumption as the tumors grow larger (Fig. 3). In animals treated with αCTLA-4, we found that the large increase in sO2 24 hours after the first dose was due to a decrease in dHb and a corresponding increase in HbO2. Specifically, there was a 28% decrease in mean dHb and a 25% increase in mean HbO2. This trend of elevated HbO2 and decreased dHb continued for the next 4 days until Day 6. Beginning Day 7, we observe an increase in dHb and a corresponding decrease in HbO2 that contributes to the decrease in sO2. On the other hand, the increase in sO2 in the tumors treated with PD-L1 were driven by change in HbO2 only, with very minor changes in dHb. Finally, both HbO2 and dHb in the COMB group showed little to no variation over the 10-day period, which explains the lack of change in mean sO2 of the COMB group.
Fig. 3. (A) Linear plots for percent change in oxygenated hemoglobin (HbO2) for αCTLA-4, αPD-L1 and COMB against the control group IgG across the 10 days of DRS. (B) Percent change in deoxygenated hemoglobin (dHb) content for the four treatment groups across 10 days of spectroscopy. Data are presented as group mean (line) ±SEM (represented by error bars). Significant differences among treatments in specific days are illustrated with black pounds (#) while significant differences of specific days within groups are represented by asterisk of their respective color indicating a statistical significance at p < 0.05.
Next, we evaluated if changes in sO2 following treatment within each group were dependent on pre-treatment, baseline sO2. We found that tumor reoxygenation measured 24 hours after each dose of ICI treatment was moderately to strongly negatively correlated with baseline vascular oxygenation (Fig. 4). While there was a very weak correlation between fold-change in sO2 on Day 2 and baseline sO2 in the PD-L1 group, this correlation became progressively stronger by Day 5 (r = -0.55) and was statistically significant on Day 8 (r = -0.75, p = 0.02). In addition, we observed a statistically significant correlation in the combined treatment group on Day 2 (r = -0.7, p = 0.05) and this negative correlation was strong across all three time points of evaluation (r = -0.66 to -0.73).
Fig. 4. Scatter plots representing the relationship between fold changes in sO2 and base line vascular oxygenation for the groups CTLA-4, PD-L1 and COMB 24 hours the respective dosage, day 2 (A), day 5 (B), and day 8 (C). Linear regressions for the respective groups are represented by the solid lines, r and p values are shown for all the groups with a significant difference at p < 0.05.
Figure 5 presents a comparison of fold changes in tissue scattering between the control group and each of the ICI treatment groups. The IgG treatment group experience a decrease in scattering values 24 h after the second ICI dosage. Although on day 5 the sudden drop is driven by a single mouse, the pattern is seen across multiple mice on days 6 and 7. However, there were no significant changes in tissue scattering in αCTLA-4 and the COMB groups over time or between the treatment groups and IgG control. The fold-change in scattering in the PD-L1 group was significantly different when compared with baseline, pre-treatment scattering (p < 0.05).
Fig. 5. Line plots of reduced scattering coefficient for αCTLA-4, αPD-L1, and αCOMB against the control group IgG (A, B, and C respectively) across 10 days of treatment monitoring with spectroscopy. Data are presented as group mean (line) ±SEM (represented by error bars). Significant differences of specific days within groups are represented by asterisk of their respective color indicating a statistical significance at p < 0.05.
The representative hematoxylin and eosin (H&E)-stained images for each group are presented in Fig. 6. While we did not see significant differences in tumor volume between the different treatment groups at the time of excision, the animals in the combined treatment group (COMB) had elevated levels of necrosis compared with the isotype control and monotherapy groups (Fig. 6(D)).
Fig. 6. Representative histopathological images of tumors from each treatment group in response to immune-checkpoint inhibitors – A. IgG isotype control, B. anti-PD-L1, C. anti-CTLA-4, and D. COMB treatment. The scale bars represent 250 µm.
4. Discussion
While previous studies have investigated the deleterious effects of hypoxia, acidification, and decreased glucose availability on immune cells and the functioning of immune checkpoint inhibitors [21, 25], studies of how immune checkpoint inhibitors affect tumor metabolism or oxygenation are limited. Here, we present the first study of functional changes within the tumor microenvironment in response to ICIs using diffuse reflectance spectroscopy. Our method provides a potentially noninvasive and nondestructive method for investigating the tumor microenvironment and evaluating treatment response to ICIs. We used CT26 colorectal tumor xenografts in mice and treated these tumors with either single ICIs or a combination and determined the longitudinal changes in tumor vascular oxygenation, hemoglobin concentration, and tissue scattering. While treatment with monotherapy and combination therapy typically elicits a complete response in some tumors [20], the lack of significant change observed here is likely due to our study design. Unlike typical animal model studies involving ICIs, which involve cancer cell inoculations followed by ICI treatment in quick succession (within 2-3 days), we waited for the development of palpable tumor masses (approx. 80-120 mm3) to ensure we could perform baseline optical spectroscopy on tumors prior to treatment.
We found a large fold-change in vascular oxygenation (sO2) or reoxygenation in tumors treated with anti-CTLA-4 and anti-PD-L1 monotherapy. The tumor growth curves for both treatment groups were identical, indicating that changes in sO2 were not driven by changes in tumor volume. However, the source of the fold-change in sO2 was different in both groups. While the reoxygenation in anti-CTLA-4 treated tumors was driven by an increase in HbO2 and a corresponding decrease in dHb, the anti-PD-L1 treated tumors only showed an increase in HbO2 and no corresponding change in dHb. These findings demonstrate that inhibition of immune checkpoints increases tumor sO2 in CT26 tumors, which points to a possible reduction in tumor hypoxia. Further work is necessary to ascertain if this reoxygenation trend is also observed in other cell lines and animal models. Previous work from our lab has shown that sO2 is negatively correlated with hypoxic fraction [10]. Changes to sO2 could stem from variation in either supply or demand. Specifically, cancer cell death and reduced oxygen consumption by tumor cells contribute to decreased utilization of oxygen from the vasculature, which can manifest as a reduction in dHb, increase in HbO2, and consequently an increase in sO2. On the other hand, increased perfusion in response to microenvironmental perturbations will manifest as an increase in HbO2, leading to an increase in sO2. Data from the isotype-treated tumors show that the progressive decrease in sO2 (increased hypoxia) is due to an increase in dHb, showing increased demand for oxygen. The corresponding increase in total hemoglobin concentration illustrates the growth of new tumor vasculature to supply the proliferating cancer cells in these untreated tumors. Tumors treated with anti-PD-L1 showed an increase in HbO2 with no change in dHb, indicating a likely increase in perfusion. Further studies on excised tumors at specific timepoints of interest in this study will be necessary to evaluate whether these perfusion-related changes were associated with or due to increased immune cell infiltration. In the first 5 days following treatment with anti-CTLA-4, the increased sO2 was due to a combination of increased HbO2 and decreased dHb. Since this did not lead to an appreciable decrease in total hemoglobin concentration, these data indicate that these changes were likely due to decreased oxygen consumption by the tumor cells. On the other hand, we found no significant changes in sO2 with respect to pre-treatment baseline in tumors treated with combination therapy and no reasonable explanation for why we did not observe a change. As indicated above, detailed studies on excised tumors or dorsal window chamber-based in vivo studies on superficial tumors could shed light on the lack of changes in vascular sO2 in tumors treated with combination ICI. An examination of the reoxygenation trends across treatment groups found that the magnitude of reoxygenation was negatively correlated with baseline sO2 values. Given our study design – injection of ICIs only after appearance of palpable tumors – we were unable to observe significant changes in tumor volumes and hence determine if baseline hypoxia contributes to treatment resistance. Elevated mitochondrial oxidative metabolism in tumor cells, which leads to intratumoral hypoxia, has been shown to decrease the efficacy of anti-PD-1 therapy in pre-clinical and clinical studies [26].
Previous work performed with an orthotopic model of CT26 tumors found a significant increase in collagen I deposition following combined treatment with anti-CTLA-4 and anti-PD-L1 [27]. However, we did not observe any significant changes in tissue scattering, which counts collagen as one of its sources, in the combined treatment group. Histological analysis of the tumor groups revealed elevated levels of necrosis in the combined treatment group compared with the monotherapy and isotype control groups. While necrotic tissue is also known to be associated with increased optical scattering [28], this was not reflected in the longitudinal changes in optical scattering. We did observe a small but significant increase in tissue scattering in the anti-PD-L1 treatment group relative to baseline. Our observations of increased tissue scattering and an increase in perfusion in response to anti-PD-L1 treatment deserves further investigation into the possibility of immune cell infiltration. While collagen, cells, nuclei, and mitochondria are considered key contributors to overall tissue scattering, the contributions of immune cell infiltration to tissue scattering and whether they are large enough to create significant changes are unknown.
5. Conclusion
In conclusion, we have demonstrated the sensitivity of diffuse reflectance spectroscopy to tumor microenvironmental changes in response to immune checkpoint inhibitors in CT26 murine colorectal cancer cells. We will note that these studies were only performed in one cell line and therefore the microenvironmental response could be different in other cell lines. A more comprehensive study involving multiple cell lines and animal models will provide a robust validation of the work presented here. Our results demonstrate a large increase in tumor oxygenation following treatment with ICIs and that the source of reoxygenation is likely dependent on the type of ICI. The magnitude of reoxygenation was strongly correlated with baseline sO2. To accurately identify the source of this reoxygenation, it will be important to also perform studies in immune-compromised mice that lack mature T cells to determine the contributions of immune cells to the functional changes observed here. The knowledge gained from these studies can be very useful in determining how to best combine such ICI treatments with other conventional therapies, such as radiation. For example, better-oxygenated tumors have been shown to respond better to radiation therapy [29]. An increase in tumor oxygenation levels, as observed here, and the timeline of such an increase following ICI treatment can provide information on how to best combine or sequence both treatments to improve treatment response rates.
Funding
National Cancer Institute (R01CA238025); Winthrop P. Rockefeller Cancer Institute, University of Arkansas for Medical Sciences (Team Science Award).
Acknowledgements
The authors would like to acknowledge Dr. Santosh Paidi and Dr. Ishan Barman at Johns Hopkins University for assistance with immunohistochemistry of tumor samples.
Disclosures
The authors declare no potential conflicts of interest.
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. M. J. Overman, R. McDermott, J. L. Leach, S. Lonardi, H. J. Lenz, M. A. Morse, J. Desai, A. Hill, M. Axelson, R. A. Moss, M. V. Goldberg, Z. A. Cao, J. M. Ledeine, G. A. Maglinte, S. Kopetz, and T. André, “Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal cancer (CheckMate 142): an open-label, multicentre, phase 2 study,” Lancet Oncol. 18(9), 1182–1191 (2017). [CrossRef]
2. D. T. Le, J. N. Uram, H. Wang, B. R. Bartlett, H. Kemberling, A. D. Eyring, A. D. Skora, B. S. Luber, N. S. Azad, D. Laheru, B. Biedrzycki, R. C. Donehower, A. Zaheer, G. A. Fisher, T. S. Crocenzi, J. J. Lee, S. M. Duffy, R. M. Goldberg, A. de la Chapelle, M. Koshiji, F. Bhaijee, T. Huebner, R. H. Hruban, L. D. Wood, N. Cuka, D. M. Pardoll, N. Papadopoulos, K. W. Kinzler, S. Zhou, T. C. Cornish, J. M. Taube, R. A. Anders, J. R. Eshleman, B. Vogelstein, and L. A. Diaz Jr., “PD-1 Blockade in Tumors with Mismatch-Repair Deficiency,” N. Engl. J. Med. 372(26), 2509–2520 (2015). [CrossRef]
3. G. T. Gibney, L. M. Weiner, and M. B. Atkins, “Predictive biomarkers for checkpoint inhibitor-based immunotherapy,” Lancet Oncol. 17(12), e542–e551 (2016). [CrossRef]
4. J. Larkin, V. Chiarion-Sileni, R. Gonzalez, J. J. Grob, C. L. Cowey, C. D. Lao, D. Schadendorf, R. Dummer, M. Smylie, and P. Rutkowski, “Combined nivolumab and ipilimumab or monotherapy in untreated melanoma,” N. Engl. J. Med. 373(1), 23–34 (2015). [CrossRef]
5. J. W. Huh, J. H. Lee, and H. R. Kim, “Prognostic significance of tumor-infiltrating lymphocytes for patients with colorectal cancer,” Arch. Surg. 147(4), 366–372 (2012). [CrossRef]
6. P.-L. Chen, W. Roh, A. Reuben, Z. A. Cooper, C. N. Spencer, P. A. Prieto, J. P. Miller, R. L. Bassett, V. Gopalakrishnan, K. Wani, M. P. De Macedo, J. L. Austin-Breneman, H. Jiang, Q. Chang, S. M. Reddy, W.-S. Chen, M. T. Tetzlaff, R. J. Broaddus, M. A. Davies, J. E. Gershenwald, L. Haydu, A. J. Lazar, S. P. Patel, P. Hwu, W.-J. Hwu, A. Diab, I. C. Glitza, S. E. Woodman, L. M. Vence, I. I. Wistuba, R. N. Amaria, L. N. Kwong, V. Prieto, R. E. Davis, W. Ma, W. W. Overwijk, A. H. Sharpe, J. Hu, P. A. Futreal, J. Blando, P. Sharma, J. P. Allison, L. Chin, and J. A. Wargo, “Analysis of immune signatures in longitudinal tumor samples yields insight into biomarkers of response and mechanisms of resistance to immune checkpoint blockade,” Cancer Discovery 6(8), 827–837 (2016). [CrossRef]
7. R. A. Drezek, M. Guillaud, T. G. Collier, I. Boiko, A. Malpica, C. E. MacAulay, M. Follen, and R. R. Richards-Kortum, “Light scattering from cervical cells throughout neoplastic progression: influence of nuclear morphology, DNA content, and chromatin texture,” J. Biomed. Opt. 8(1), 7–17 (2003). [CrossRef]
8. F. Hu, H. Martin, A. Martinez, J. Everitt, A. Erkanli, W. T. Lee, M. Dewhirst, and N. Ramanujam, “Distinct angiogenic changes during carcinogenesis defined by novel label-free dark-field imaging in a hamster cheek pouch model,” Cancer Res. 77(24), 7109–7119 (2017). [CrossRef]
9. D. G. Johnston, “Cytoplasmic: Nuclear ratios in the cytological diagnosis of cancer,” Cancer 5(5), 945–949 (1952). [CrossRef]
10. S. Dadgar, J. R. Troncoso, and N. Rajaram, “Optical spectroscopic sensing of tumor hypoxia,” J. Biomed. Opt. 23(6), 067001 (2018). [CrossRef]
11. G. M. Palmer, R. J. Viola, T. Schroeder, P. S. Yarmolenko, M. W. Dewhirst, and N. Ramanujam, “Quantitative diffuse reflectance and fluorescence spectroscopy: tool to monitor tumor physiology in vivo,” J. Biomed. Opt. 14(2), 024010 (2009). [CrossRef]
12. S. Dadgar, J. R. Troncoso, E. R. Siegel, N. M. Curry, R. J. Griffin, R. P. M. Dings, and N. Rajaram, “Spectroscopic investigation of radiation-induced reoxygenation in radiation-resistant tumors,” Neoplasia 23(1), 49–57 (2021). [CrossRef]
13. P. M. Diaz, S. V. Jenkins, K. Alhallak, D. Semeniak, R. J. Griffin, R. P. M. Dings, and N. Rajaram, “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). [CrossRef]
14. F. Hu, K. Vishwanath, J. K. Salama, A. Erkanli, B. Peterson, J. R. Oleson, W. T. Lee, D. M. Brizel, N. Ramanujam, and M. W. Dewhirst, “Oxygen and perfusion kinetics in response to fractionated radiotherapy in FaDu head and neck cancer xenografts are related to treatment outcome,” Int. J. Radiat. Oncol., Biol., Phys. 96(2), 462–469 (2016). [CrossRef]
15. K. Vishwanath, D. Klein, K. Chang, T. Schroeder, M. W. Dewhirst, and N. Ramanujam, “Quantitative optical spectroscopy can identify long-term local tumor control in irradiated murine head and neck xenografts,” J. Biomed. Opt. 14(5), 054051 (2009). [CrossRef]
16. U. Sunar, H. Quon, T. Durduran, J. Zhang, J. Du, C. Zhou, G. Yu, R. Choe, A. Kilger, R. Lustig, L. Loevner, S. Nioka, B. Chance, and A. G. Yodh, “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). [CrossRef]
17. L. Dong, M. Kudrimoti, D. Irwin, L. Chen, S. Kumar, Y. Shang, C. Huang, E. L. Johnson, S. D. Stevens, B. J. Shelton, and G. Yu, “Diffuse optical measurements of head and neck tumor hemodynamics for early prediction of chemoradiation therapy outcomes,” J. Biomed. Opt. 21(8), 085004 (2016). [CrossRef]
18. L. Dong, M. Kudrimoti, R. Cheng, Y. Shang, E. L. Johnson, S. D. Stevens, B. J. Shelton, and G. Yu, “Noninvasive diffuse optical monitoring of head and neck tumor blood flow and oxygenation during radiation delivery,” Biomed. Opt. Express 3(2), 259–272 (2012). [CrossRef]
19. M. J. Selby, J. J. Engelhardt, R. J. Johnston, L.-S. Lu, M. Han, K. Thudium, D. Yao, M. Quigley, J. Valle, C. Wang, B. Chen, P. M. Cardarelli, D. Blanset, and A. J. Korman, “Preclinical development of ipilimumab and nivolumab combination immunotherapy: mouse tumor models, in vitro functional studies, and cynomolgus macaque toxicology,” PLoS One 11(9), e0161779 (2016). [CrossRef]
20. J. Duraiswamy, K. M. Kaluza, G. J. Freeman, and G. Coukos, “Dual blockade of PD-1 and CTLA-4 combined with tumor vaccine effectively restores T-Cell rejection function in tumors,” Cancer Res. 73(12), 3591–3603 (2013). [CrossRef]
21. C.-H. Chang, J. Qiu, D. O’Sullivan, M. D. Buck, T. Noguchi, J. D. Curtis, Q. Chen, M. Gindin, M. M. Gubin, and G. J. Van Der Windt, “Metabolic competition in the tumor microenvironment is a driver of cancer progression,” Cell 162(6), 1229–1241 (2015). [CrossRef]
22. B. S. Nichols, N. Rajaram, and J. W. Tunnell, “Performance of a lookup table-based approach for measuring tissue optical properties with diffuse optical spectroscopy,” J. Biomed. Opt. 17(5), 057001 (2012). [CrossRef]
23. N. Rajaram, T. Nguyen, and J. Tunnell, “Lookup table-based inverse model for determining optical properties of turbid media,” J. Biomed. Opt. 13(5), 050501 (2008). [CrossRef]
24. N. Rajaram, A. Gopal, X. Zhang, and J. Tunnell, “Experimental validation of the effects of microvasculature pigment packaging on in vivo diffuse reflectance spectroscopy,” Lasers Surg. Med. 42(7), 680–688 (2010). [CrossRef]
25. N. E. Scharping, A. V. Menk, R. D. Whetstone, X. Zeng, and G. M. Delgoffe, “Efficacy of PD-1 blockade is potentiated by Metformin-induced reduction of tumor hypoxia,” Cancer Immunol. Res. 5(1), 9–16 (2017). [CrossRef]
26. Y. G. Najjar, A. V. Menk, C. Sander, U. Rao, A. Karunamurthy, R. Bhatia, S. Zhai, J. M. Kirkwood, and G. M. Delgoffe, “Tumor cell oxidative metabolism as a barrier to PD-1 blockade immunotherapy in melanoma,” JCI Insight4 (2019). [CrossRef]
27. E. Fiegle, D. Doleschel, S. Koletnik, A. Rix, R. Weiskirchen, E. Borkham-Kamphorst, F. Kiessling, and W. Lederle, “Dual CTLA-4 and PD-L1 blockade inhibits tumor growth and liver metastasis in a highly aggressive orthotopic mouse model of colon cancer,” Neoplasia 21(9), 932–944 (2019). [CrossRef]
28. K. Vishwanath, H. Yuan, W. T. Barry, M. W. Dewhirst, and N. Ramanujam, “Using optical spectroscopy to longitudinally monitor physiological changes within solid tumors,” Neoplasia 11(9), 889–900 (2009). [CrossRef]
29. D. Brizel, G. Sibley, L. Prosnitz, R. Scher, and M. Dewhirst, “Tumor hypoxia adversely affects the prognosis of carcinoma of the head and neck,” Int. J. Radiat. Oncol., Biol., Phys. 38(2), 285–289 (1997). [CrossRef]
