Photochemistry and Photobiology. 2023;00:1–18. | 1wileyonlinelibrary.com/journal/php Received: 10 May 2023 | Revised: 23 June 2023 | Accepted: 26 June 2023 DOI: 10.1111/php.13836 S P E C I A L I S S U E R E S E A R C H A R T I C L E Test method for evaluating the photocytotoxic potential of fluorescence imaging products Shruti Vig1 | Brandon Gaitan1,2 | Lucas Frankle1 | Yu Chen3 | Rosalie Elespuru2 | T. Joshua Pfefer2 | Huang- Chiao Huang1 1Fischell Department of Bioengineering, University of Maryland, College Park, Maryland, USA 2Center for Devices and Radiological Health, US Food and Drug Administration, Silver Spring, Maryland, USA 3Department of Biomedical Engineering, University of Massachusetts, Amherst, Massachusetts, USA Correspondence Huang- Chiao Huang, Fischell Department of Bioengineering, University of Maryland, College Park, MD, USA. Email: hchuang@umd.edu Funding information NSF- FDA, Grant/Award Number: 2037815; MPower Fellowship; Clark Doctoral Fellowship; University of Maryland ASPIRE Fellowship Abstract Various fluorescence imaging agents are currently under clinical studies. Despite significant benefits, phototoxicity is a barrier to the clinical translation of fluoro- phores. Current regulatory guidelines on medication- based phototoxicity focus on skin effects during sun exposure. However, with systemic and local adminis- tration of fluorophores and targeted illumination, there is now possibility of pho- tochemical damage to deeper tissues during intraoperative imaging procedures. Hence, independent knowledge regarding phototoxicity is required to facilitate the development of fluorescence imaging products. Previously, we studied a cell- free assay for initial screening of reactive molecular species generation from fluo- rophores. The current work addresses a safety test method based on cell viability as an adjunct and a comparator with the cell- free assay. Our goal is to modify and implement an approach based on the in vitro 3T3 neutral red uptake assay of the Organization for Economic Co- Operation and Development Test Guideline 432 (OECD TG432) to evaluate the photocytotoxicity of clinically relevant fluo- rophores. These included indocyanine green (ICG), proflavine, methylene blue (MB), and IRDye800, as well as control photosensitizers, benzoporphyrin deriva- tive (BPD) and rose bengal (RB). We performed measurements at agent concen- trations and illumination parameters used for clinic imaging. Our results aligned with prior studies, indicating photocytotoxicity in RB and BPD and an absence of reactivity for ICG and IRDye800. DNA interactive agents, proflavine and MB, exhibited drug/light dose– response curves like photosensitizers. This study pro- vides evidence and insights into practices useful for testing the photochemical safety of fluorescence imaging products. K E Y W O R D S fluorescence imaging, fluorescence imaging agents, neutral red uptake (NRU) assay, OECD TG432, photocytotoxicity This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2023 The Authors. Photochemistry and Photobiology published by Wiley Periodicals LLC on behalf of American Society for Photobiology. This article is part of a Special Issue dedicated to the topic of Critical Issues and Recent Progresses in Clinical Translation of Photomedicine. Abbreviations: BPD, benzoporphyrin derivative; FBS, fetal bovine serum; FGS, fluorescence- guided surgery; ICG, indocyanine green; MB, methylene blue; NRU, neutral red uptake; PDT, photodynamic therapy; RB, rose bengal; RMS, reactive molecular species. www.wileyonlinelibrary.com/journal/php https://orcid.org/0000-0002-0534-1571 mailto: https://orcid.org/0000-0002-5406-0733 mailto:hchuang@umd.edu http://creativecommons.org/licenses/by-nc-nd/4.0/ http://crossmark.crossref.org/dialog/?doi=10.1111%2Fphp.13836&domain=pdf&date_stamp=2023-07-26 2 | PHOTOCHEMISTRY AND PHOTOBIOLOGY INTRODUCTION Fluorescence imaging can aid in the visualization of crit- ical nerves, vascular structures, tumor margins, and me- tastases. This optical tool is minimally invasive, real- time, and provides a better resolution for detection compared to naked eye, and thus is being increasingly accepted by cli- nicians.1,2 Fluorescent agents can be broadly classified as non- targeted, metabolic, and molecular- targeted agents.3 For example, indocyanine green (ICG) and fluorescein are non- targeted fluorescent agents used for retinal an- giography4,5 and fluorescence- guided surgery (FGS).6,7 Methylene blue (MB) has been studied for tumor imag- ing8– 11 and sentinel lymph node mapping for axillary stag- ing in breast cancer.12,13 Proflavine's ability to bind to cell nuclei offers a proper assessment of nuclear architecture and grading of oral carcinoma.14 Molecular- targeted ap- proaches have also been explored to improve the selec- tivity of fluorescent agents. For example, conjugation of IRDye800CW to tumor- targeting moieties, such as an- tibodies, is under clinical investigations for FGS of head and neck cancer15 and visualization of other cancers.16 Many fluorescent agents can also generate cytotoxic reac- tive molecular species (RMS), making them ideal for com- bined optical imaging and phototherapy applications. For instance, MB is being evaluated for real- time visualization of ureters during abdominal and laparoscopic surger- ies,17,18 identification of enlarged thyroid and parathyroid glands,19,20 as well as fluorescence imaging and photo- therapy of localized cutaneous amyloidosis and toenail onychomycosis.21– 23 ICG has also been studied for pho- todynamic therapy (PDT) and is currently being assessed for usage in periodontal disease and with diabetic patients having peri- implantitis.24,25 The safe use of light- emitting devices for medical ap- plications is addressed by well- established international consensus standards.26,27 Yet, the use of such devices in combination with fluorophores is not satisfactorily ad- dressed in these documents. One potential barrier to the clinical translation of products involving optical devices and fluorophores is normal tissue phototoxicity. When exposed to light at specific wavelengths, fluorophores can produce photochemical reactions that cause phototoxic- ity.28 These photoreactive agents often undergo electronic excitation and produce RMS like singlet oxygen, super- oxide, hydroxyl radicals, and hydrogen peroxide, which can cause cellular and tissue damage through Type I and Type II photochemical reactions.29 Type I and II reactions require oxygen as a reagent for the formation of RMS.30 Type I reactions lead to the formation of hydroxyl radicals, hydrogen peroxide, and superoxide ions. Type II reactions involve an energy transfer reaction between the photoacti- vated agent and molecular oxygen, leading to the formation of singlet oxygen. Thus, given the fluorophore potential to cause unintentional damage to a patient through various mechanisms, it is important to evaluate their phototox- icity to ensure safe clinical application. Independent of the nature of toxicity generated by the photoactivation of these fluorophores, the outcomes are cellular alterations or lethality. While studies have focused on developing test methods for detecting phototoxicity, there is still a dearth of standardized test methods for the photochemical safety of clinical fluorescence imaging products. The neutral red (3- amino- 7- dimethyl- 2- methyl phenazine hydrochloride) uptake (NRU) assay was de- veloped in 1985 by Borenfreund and Puerner, as a cell viability assay based on the ability of cells to incorporate and bind to neutral red dye (NR).31 NR is a weak cationic dye with a net charge close to zero. It penetrates cells by non- ionic diffusion at physiological pH and accumulates intracellularly in lysosomes.32 Inside the lysosomes exists a proton gradient, due to which the dye becomes charged and is retained. The NRU assay allows the assessment of membrane permeability and lysosomal activity, mak- ing it possible to differentiate between viable, damaged, and dead cells. The assay was an early attempt to estab- lish a more quantitative, reproducible, inexpensive, and sensitive method for in vitro toxicity screening compared to other subjective methods (like morphological inspec- tion).31 The assay has been adapted to assess cytotoxic- ity and phototoxicity screening of chemicals/products in vitro.33,34 Several validation studies have been set up for the NRU assay.35,36 In 2000, the NRU test on BALB/c 3T3 mouse fibroblasts to assess phototoxicity was accepted in the EU. In 2004, the Organization for Economic Co- operation and Development (OECD) described a method to evaluate photocytotoxicity in the OECD TG 432 guide- line.37 In 2013, EURL ECVAM (European Commission Joint Research Centre) published regulatory recommen- dations on the use of the 3T3 NRU assay to predict acute oral toxicity of chemicals.38 The Center for Drug Evaluation and Research (CDER) and the Center for Biologics Evaluation and Research (CBER) of the U.S. Food and Drug Administration (FDA) recommend phototoxic assessment of agents that absorb light within the range of natural sunlight (290– 700 nm).39 The International Council for Harmonization of technical requirements for pharmaceuticals for human use (ICH) S10 guidelines emphasize the need to develop initial in vitro phototoxicity tests before carrying out clinical trials. ICH S10 provides guidelines on the initial photochemical properties of agents to be considered for phototoxic test- ing. For an agent to demonstrate potential phototoxicity, it should (a) absorb light within the wavelength range of 290– 700 nm (sunlight spectrum), (b) generate RMS fol- lowing absorption of UV– visible light, and (c) showcase 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 3VIG et al. a molar extinction coefficient greater than 1000 L/mol/ cm. FDA currently accepts the in vitro 3T3 NRU assay for evaluating the phototoxicity of drugs.37 OECD TG432 guidelines recommend using a solar simulator to mimic sunlight exposure at a set non- cytotoxic irradiation dose of 5 J/cm2 at 1.7 mW/cm2 as an acceptable light source for 3T3- NRU phototoxicity testing. While photosensitiv- ity due to solar exposure has historically been a primary concern, there is now the potential for photochemical damage to internal organs and tissues during intraopera- tive imaging procedures. Furthermore, the emerging NIR fluorophores absorb irradiations at high fluence targeted wavelengths that may be beyond the sunlight spectrum. Thus, independent knowledge regarding RMS generation and related photocytotoxicity for existing and emerging fluorescent dyes at doses and exposure levels used for clin- ical imaging is needed to streamline product development and regulations for clinical applications. In this study, we used a modified phototoxicity assess- ment strategy based on OECD TG432 guidelines to assess the photocytotoxicity of contrast- enhanced fluorescence imaging products for a range of concentrations and illu- minations within and above the clinical range used for diagnostic imaging. Instead of using solar radiation for these experiments, the NRU assay was performed at con- centration ranges and illumination parameters used in the clinic for diagnostic imaging to establish a dose– response curve for each contrast agent. As outlined in the OECD TG432 guidance document, mean photo effect (MPE) was calculated for fluorophore concentration ranges at tested radiant exposures (He) [(J/cm2)], which is defined as a product of laser irradiance and time. In addition to pro- viding insights into assessing photocytotoxicity generated by fluorescence imaging products and the limitations of the assay approach, this study highlights potential best practices for preclinical safety assessment of contrast- enhanced fluorescence imaging products. MATERIALS AND METHODS Overview In this study, the 3T3 NRU phototoxicity assessment37 was modified by: (a) expanding the recommended illumi- nation wavelength range beyond 290– 700 nm to accom- modate long wavelength contrast agents for phototoxicity testing; (b) establishing the dose– response curve for agent concentrations from available literature and OECD- recommended concentrations to account for variance in photocytotoxicity to determine potential safety thresholds for fluorophores; (c) evaluating photocytotoxicity gener- ated at clinically used illumination levels— excitation wavelengths, irradiance, and radiant exposure (He) using laser diodes, as opposed to irradiation with a broadband solar simulator suggested in current guidelines. The study encompassed four phases (1) utilize agent concentrations, optical exposure wavelengths, and He levels from avail- able literature on clinical and animal studies summarized in our previous study and Table 1; (2) obtain experimen- tal data with each fluorophore using the NRU assay; (3) analyze the resultant dose– response curves to assess prediction model values (MPE) for different fluorophore products; and (4) compare the results with published lit- erature and assess overall performance and limitations of the assay. The existing method for testing medication- based phototoxicity suggested by the OECD guidelines is summarized in Figure S1. Cell culture BALB/c 3T3 clone A31 (ATCC) cells were cultured follow- ing recommendations from the OECD 432 guidelines.37 Briefly, cells were cultured in T- 75 flasks (Cell Treat) in a 37°C and 5% CO2 incubator until they reached 80%– 90% confluency. Dulbecco's modified eagle's medium (DMEM) (ATCC) supplemented with 10% fetal bovine serum (FBS) (Gibco), 1% penicillin– streptomycin (Corning) were used to maintain the cells. The medium was changed every 48 h. BALB/c 3T3 cells were confirmed to be mycoplasma free using the Mycoalert™ Plus mycoplasma detection kit (Lonza). To optimize the cell seeding for phototoxicity evaluation, BALB/c 3T3 cells were plated over a range of cell seeding numbers (from 1.0 × 103 to 1.2 × 105 cells per well) in 96- well tissue culture plates (Southern Labware). After 48 h, an MTT (3- (4,5- dimethylthiazol- 2- yl)- 2,5- diphe nyltetrazolium bromide) assay (Invitrogen) was performed following the vendor protocol. Briefly, 3T3 cells were in- cubated with 0.25 mg/mL of MTT for 1 h. The formazan crystals formed were solubilized in 100% DMSO, and the absorbance was measured at 570 nm using a spectropho- tometer (Synergy Neo2; Biotek) (Figure S2). Fluorophore concentrations and illumination levels RMS generatiosn by contrast agent depends on the ex- citation spectrum and He of the illumination source in addition to the concentration of the agent. This study focused on the phototoxic potential of fluorescence im- aging products at clinically relevant concentrations and exposure parameters. The literature review identified appropriate parameters for testing each contrast agent. Fluorophores were tested for concentrations ranging from 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4 | PHOTOCHEMISTRY AND PHOTOBIOLOGY T A B L E 1 Li te ra tu re re vi ew su m m ar y on th e cl in ic al u se o f i m ag in g flu or op ho re s t es te d in th e st ud y: c on ce nt ra tio n, e xc ita tio n w av el en gt h, a nd ir ra di an ce . Fl uo ro ph or e M od el U se c as e C on ce nt ra ti on E xc it at io n w av el en gt h (n m ) Ir ra di an ce Im ag in g ti m e C it at io n IC G H um an Ly m ph at ic im ag in g 0. 2 m L in je ct ed 76 0 N A N A U nn o et a l.81 H um an Ly m ph at ic im ag in g 1 m L of 0 .5 % IC G in je ct ed 75 0– 80 0 N A 30 m in Ta ke uc hi e t a l.82 H um an Bl oo d co nc . o f I C G 5– 30 m g/ L in p la sm a 80 5 N A N A Im ai e t a l.83 N A In st ru m en t r ev ie w N A 78 5 1. 9 m W /c m 2 N A Zh u et a l.42 H um an Ly m ph n od e im ag in g 1 m L in je ct ed 76 0 4 m W /c m 2 15 m in Ta ga ya e t a l.41 IR D ye 80 0 H um an H ea d an d ne ck su rg er y 5. 2– 13 0 m g in je ct ed 77 5 N A N A G ao e t a l.84 H um an G lio bl as to m a im ag in g 50 – 1 00 m g in je ct ed N A N A N A M ill er e t a l.45 N A In st ru m en t r ev ie w N A 78 0 3– 30 m W /c m 2 N A D 'S ou za e t a l.46 Pr of la vi ne H um an O ra l c an ce r i m ag in g 0. 01 % (w /v ) 45 5 N A N A Sh in e t a l.48 H um an O ra l c an ce r i m ag in g 0. 01 % 45 5 N A 3– 15 m in 85 H um an Im ag in g of B ar re tt' s r el at ed ne op la si a 0. 01 % (w /v ) 43 5/ 50 0 8. 2 an d 14 .9 m W N A Ta ng e t a l.47 M B H um an In tr av en ou s b io - a va ila bi lit y 60 0– 20 00 n g/ m L in b lo od N A N A N A St ef an o et a l.43 Sw in e U te ri ne im ag in g 0. 1 m g/ kg in je ct ed 67 0 2. 5 m W /c m 2 N A M at su i e t a l.44 H um an U te ri ne im ag in g 0. 25 – 1 m g/ kg in je ct ed 67 0 1. 08 m W /c m 2 > 5 m in V er be ek e t a l.40 H um an Br ea st c an ce r i m ag in g 1 m g/ kg in je ct ed 67 0 1. 08 m W /c m 2 N A Tu m m er s e t a l.10 A bb re vi at io ns : I C G , i nd oc ya ni ne g re en ; M B, m et hy le ne b lu e. So ur ce : A da pt ed fr om o ur p re vi ou s s tu dy .56 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 5VIG et al. 0 μM (negative control) to approximately twice the con- centration used in the clinic for imaging. A brief litera- ture review summary on the illumination parameters of the contrast agents can be found in Table 1. The exposure time depends on acquiring adequate fluorescence signal during intraoperative imaging of tissues.40,41 The 2× con- centration range used for each agent was based on param- eters reported in the literature, like human blood plasma levels, pharmacokinetics data after intravenous injec- tion, and clinical studies on using the test agents for FGS. The 2× ICG concentration range was based on the blood plasma values found in human subjects. The illumination parameters (He and excitation wavelength) for ICG were derived from values used for fluorescent- guided surgery.42 The concentration range of 1– 7.5 μM for MB was used based on the pharmacokinetics data acquired after oral dose of 100 mg in human subjects.43 The He (0– 6 J/cm2) was determined through camera exposure time required during fluorescence- guided identification of ureters using MB.44 The 2.5– 20 μM concentration range for IRDye800 was based on clinical studies using IRDye800 as contrast agent for FGS.45,46 The He range was based on current imaging devices and the likely duration of imaging dur- ing surgery.46 For proflavine, concentration range of 1.5– 12 μM, He range of 0– 6 J/cm2, and excitation wavelength of 445 nm were based on human trials using proflavine as an imaging agent to detect oral cancer.47,48 Rose bengal (RB) and BPD were included as positive controls since they are known to generate RMS after photoactivation and are used as photodynamic agents.49,50 The BPD con- centration range of 0– 10 μM and He was based on in vivo studies where 0.5 mg of the agent was intravenously ad- ministered in rabbits.51 RB concentrations (0– 10 μM) and exposure levels (He, 0.2 J/cm2) were selected to achieve similar dose levels as BPD to compare results. The con- centration range of test agents was increased up to 100 μg/ mL per OECD 432 guidelines to test the photocytotoxicity generated at higher concentrations to determine potential safety thresholds for the fluorophores. The He and illumi- nation wavelengths were determined from clinical stud- ies using contrast agents for FGS, camera exposure times, and imaging duration during surgery. The concentration ranges, excitation wavelengths, irradiance, and He used in the study are summarized in Table 2. Optical exposure system setup and validation A custom setup was developed to illuminate 4 wells si- multaneously in a 2 × 2 square format within a 96- well plate (Figure 1). The bottom- up exposure setup comprised a laser diode controller (Thorlabs, Inc.), a laser diode, an achromatic collimating lens (f = 50 mm), a 45° mirror, and a 20° square diffuser (ED1- S50- MD) (Thorlabs, Inc.). The laser light was collimated, reflected 90°, passed through a diffuser, and projected onto a 0.5″ thick black marine board plate (Onlinemetals) with a 1″ × 1″ aperture to allow illu- mination of the four wells. Laser diodes (Thorlabs, Inc.) with the following central wavelengths were used: 520 nm (RB, LP520- SF15), 785 nm (ICG/IRDye800, L785H1), 660 nm (MB, L660P120), 685 nm (BPD, HL6750MG), and 450 nm (PL450B, Proflavine). Power measurements were taken at the set current (mA) used during treatment to ensure uniform power distribution during laser exposure. An actuated iris diaphragm (Thorlabs, SM1D12) was at- tached to the power meter sensor (model PMD100D, Thorlabs, Inc.) sized at 1:1 (area of diaphragm:active area of the power meter sensor— 1.31 cm2) and placed above the 1″ × 1″ aperture to measure the output power at each of the four well locations. The location of each well dur- ing illumination was marked on the board plate to ensure that measurements were taken from a fixed location. Irradiance (mW/cm2) was calculated by dividing the laser diode power readings by the active sensor area. The same procedure was carried out on all the laser diodes. T A B L E 2 Summary of the concentration range, excitation wavelength (laser diodes), irradiance, and He (maximum tested) values used in our study for the modified NRU assay. Fluorophore Extended conc. range (μM)- OECD Conc. range (μM)- imaging (2×) Excitation wavelength (nm) Irradiance (mW/cm2) Exposure time (s) Max tested radiant exposure, He (J/cm2) ICG 0– 175 6.5– 26 785 10 600 6 IRDYE800 0– 75 2.5– 20 785 10 600 6 Proflavine 0– 1000 0.5– 12 450 4.2 143 0.6 MB 0– 500 1– 7.5 660 10 600 6 Benzoporphyrin derivative 0– 10 0.5– 1 685 7.2 28 0.2 Rose bengal 0– 10 0.5– 2 520 1.6 125 0.2 Abbreviations: ICG, indocyanine green; MB, methylene blue. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 6 | PHOTOCHEMISTRY AND PHOTOBIOLOGY Modified neutral red uptake (NRU) assay For testing, 1.0 × 104 BALB/c 3T3 A31 clone cells were plated in 100- μL complete growth medium per well in two 96- well, black- walled, clear flat bottom microtiter plates (Southern Labware). The cells were allowed overnight attachment at 37°C in a humidified atmosphere contain- ing 5% CO2. The 96- well plates were labeled as irradiation (IRR) or NO IRR. On Day 1, a fresh fluorophore solution was prepared in DMSO or HBSS based on the solubility of the agent. Eight different testing concentrations were prepared per fluorophore. Each testing concentration had six replicates per plate (Figure S3 for the sample plating layout). For testing, the wells were washed with 100 μL of HBSS, followed by adding 100 μL of the test fluorophore concentration, and incubated for 60 min in both plates. The no- treatment control wells contained 100 μL of the solvent control (0.1% DMSO in HBSS or HBSS only). After 60 min, the IRR plate was illuminated to achieve the de- sired He at set irradiance (Table 2). For He of 1 J/cm2, the wells were irradiated in 60/4 radiations (total wells/num- ber of wells illuminated with the 2 × 2 exposure area at a time). Exposure time for the IRR group varied depending on the test He for each agent. The NO IRR plate was stored in the dark at room temperature while illuminating the IRR plate. For testing at higher He (6 J/cm2), the experi- ment was carried out using four instead of two 96- well plates. To minimize the exposure of cells to room tem- perature for prolonged time intervals (<35 min) during illumination, (i) two instead of eight fluorophores con- centrations, with dark and solvent controls, were tested per plate; (ii) four replicates per condition were tested at higher He. For He of 6 J/cm2, 12 of 24 wells (IRR group only) per plate were irradiated in 12/4 irradiations. After illumination, the solution was removed, and the cells were washed twice with 200 μL of HBSS and incubated with the cell growth medium for 24 h. The NRU assay was per- formed on Day 2 using the neutral red assay kit (Abcam). NRU assay is based on detecting viable cells via the uptake of the dye- neutral red that stains lysosomes in viable cells. Briefly, treated cells were incubated with 150 μL of the 1× neutral red solution for 3 h based on the parameters opti- mized in a previous study.52 The cells were washed with 200 μL of the wash solution (1× PBS). Consequently, cells are incubated with 200 μL of solubilization solution to release the incorporated neutral red dye under acidified- extracted conditions. The absorbance of NR dye in the solubilization solution was measured at 540 nm using a spectrophotometer (Synergy Neo2; BioTek). The modi- fied method for photocytotoxicity assessment of fluoro- phore products is summarized in Figure 2. Surviving cells were normalized to the no treatment (solvent controls) on each plate. Further analysis on endpoint readout of NRU assay is depicted in Figure S3. To determine if the absorb- ance spectrum of the test fluorophore affects the NRU ab- sorbance spectrum during phototoxicity evaluation, the cells were incubated with the fluorophores on Day 1, for about 1.5 h, considering the time for fluorophore incuba- tion and illumination. Control wells were treated with 100 μL of HBSS. After incubation, 3T3 cells were washed with HBSS and incubated with media overnight. For the fluorophores- alone group, cells were incubated with fluo- rophores overnight. As described previously, the NRU assay was performed on Day 2, and the absorbance spec- trum was recorded using the spectrophotometer (Synergy Neo2; BioTek). F I G U R E 1 Schematic of the laser exposure system. A laser diode was coupled to a collimating lens; the beam was then reflected at 90° using a mirror and passed through a 20° diffuser. The light was subsequently passed through a 1″ × 1″ square aperture, where it exposed the bottom of a 96- well plate. The top view shows the irradiation format of the samples plated on the 96- well plate. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 7VIG et al. Data analysis The concentration- dependent cytotoxic responses in the presence of light (IRR) and in the absence of light (NO IRR) were determined in the NRU assay. For data evaluation, the prediction value recommended in OECD 432, the MPE, was calculated using Phototox version 2.0 software.37 The MPE prediction model was proposed by Holzhütter et al.53 and has been used for the phototox- icity assessment of chemical agents.54 MPE is calculated based on a comparison of the IRR and NO IRR response curves on a set of commonly tested concentration ranges (i = 1, …, n) of IRR and NO IRR curves (Equation 1). (1) MPE = ∑n i=1 wiPEci ∑n i=1 wi F I G U R E 2 Modified test method for assessing phototoxicity of fluorophore products activated by lasers at specific wavelengths and illumination parameters. BALB/3T3 cells are exposed to different concentrations of the fluorophore of interest. To study the effect of tissue phototoxicity caused during diagnostic imaging, the 96- well plate is exposed to 2× illumination parameters used clinically for fluorophore excitation. NRU assay is performed on Day 2, and the absorbance values of uptake NR dye are recorded at 540 nm using a plate reader. IRR, irradiation; NRU, neutral red uptake. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 8 | PHOTOCHEMISTRY AND PHOTOBIOLOGY where photo- effect (PEc) at any concentration C is a product of dose- effect (DEc) and the response effect (REc). The dose- effect (DEc) was calculated as C* represents the concentration at which the IRR re- sponse equals the NO IRR response at concentration C. If C* cannot be determined because the response of the IRR curve is higher or lower than the NO IRR, the DE is set to 1 at that concentration. The response effect (REc) at con- centration C represents the difference between responses observed in the absence and presence of light, that is, The weighing factors wi were determined by the highest response value, that is, wi = MAX [Ri (IRR), Ri(NO IRR)]. The MPE was subsequently obtained by averaging across all PEc values. For calculating MPE using the Phototox software, the set of concentration- response values was input into the software. Fitting the curve to the data by non- linear regression was performed using the Phototox V.2 software. A bootstrap resampling (i.e., selecting data points randomly with replacements) was performed to as- sess the influence of data variability on the fitted curve. Based on the MPE values, the fluorophore product at each tested He level was estimated to be non- phototoxic, equivocal- phototoxic, or phototoxic, using the standard in OECD 432. An example of the detailed calculation for es- timating MPE values can be found in Figures S4 and S5. Statistical analysis The cell viability response was calculated using the nor- malized average absorbance values of replicates obtained using the NRU assay. The cell viability was presented as mean ± standard error of the mean (SEM). Statistical analysis was conducted using GraphPad PRISM version 9.0.2. Data were analyzed using a paired Student's t- test. A value of p ≤ 0.05 was considered statistically significant. All experiments were repeated at least thrice to assess re- peatability (n ≥ 3). RESULTS Validation of illumination system and NRU assay To investigate variation in irradiance during the illu- mination experiments for photocytotoxicity evaluation, irradiance of the laser diode at the 2 × 2 illumination area was recorded. Figure  3 shows the average irradiance re- corded using the power sensor during the illumination of wells. The actuated iris diaphragm, sized at 1:1, was placed above the four open exposure areas at each well location. The average power distribution among the four well- open areas exposed to the laser was consistent for the adjacent wells during treatment. The variance in output was calcu- lated to be less than 10% between different repeats during illumination experiments. NRU assay involves absorbance measurement of neu- tral red as the endpoint to quantify cell viability. Since the test fluorophores are also dyes, we investigated the DEc = |||| C∕C∗ − 1 C∕C∗ + 1 |||| REc = [ Rc (NO IRR) − Rc(IRR) ] ∕100 F I G U R E 3 Laser exposure system validation for used laser diodes. Individual average irradiance readings of four wells exposed to laser diode through the 1″ × 1″ aperture for (A) 660nm, (B) 520nm, (C) 450nm, (D) 685nm, and (E) 785nm laser diodes; plots show less than 10% variation in irradiance among the wells (n = 3). 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 9VIG et al. interference of fluorophore absorbance with the end- point NR dye absorbance. In Figure  4A– F, we present the absorbance spectra of the NR dye after the treatment of cells with a test fluorophore. The absorption peak of NR dye can be seen at 540 nm in the wells treated with HBSS only [NRU(3T3 + HBSS)]. There was minimal to no crossover with the NR dye for most fluorophores tested, such as ICG, IRDye800, MB, and BPD. The exception was RB, MB, and Proflavine (to some extent). To further assess whether the residual fluorophore in the wells (due to fluorophore uptake or incomplete HBSS washes) affected the NR dye absorption reading, the absorbance spectra of NR dye were recorded post- fluorophore incu- bation [NRU (3T3 + Test Fluorophore)]. No change in the NR absorbance spectrum was observed due to flu- orophore incubation between the NRU (3T3 + HBSS) F I G U R E 4 Comparative absorbance spectra of NR dye with tested fluorophores. Absorbance spectra were measured for (A) RB, (B) BPD, (C) Proflavine, (D) MB, (E) IRDye800, and (F) ICG. Plots demonstrate the peak- normalized absorbance of tested fluorophores incubated with 3T3 cells, NR dye post- incubation of cells with fluorophores, and NR dye post- incubation with solvent. NRU: neutral red uptake. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 10 | PHOTOCHEMISTRY AND PHOTOBIOLOGY and the NRU (3T3 + Test Fluorophore) groups for all the test fluorophores. Thus, it was concluded that the fluo- rophore absorption overlap with NR dye does not affect the NR absorbance in the study. Evaluation of the photocytotoxicity using the modified NRU assay The cell viability curves were obtained using the NRU assay at test concentrations, clinically relevant excita- tion wavelengths, and He for all the tested fluorophores in this study. Dose– response curves for the positive controls at the extended concentration ranges recom- mended by OECD for solar light- activated studies, along with 2× of clinically used concentrations, are shown in Figure  5. For the positive control, RB, a concentration- dependent decrease in cell viability was observed at the He of 0.2 J/cm2. A concentration- dependent decrease in cell viability was also observed beyond 1 μM in the NO IRR group due to the intrinsic toxicity of the RB.55 The agent shows light- independent cytotoxicity in the 1– 10 μM dose range. A similar expected trend was observed for BPD products at He as low as 0.2 J/cm2, indicating significant F I G U R E 5 Dose– response curves of positive controls. Rose bengal and BPD results for clinically relevant concentrations (shaded in gray) are shown in (A) and (C), respectively. Corresponding results for the OECD TG432 recommended range are shown in (B) and (D). The shaded region in the plots is a comparative representation of the clinically relevant 2× dose range of the test agent within the OECD recommend concentration range. All data points are average of three repeats. The error bars represent the standard error of mean (SEM). BPD, benzoporphyrin derivative derivative; IRR, irradiation. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 11VIG et al. photocytotoxicity in a concentration- dependent manner. However, no non- irradiation toxicity was observed for BPD at the tested range. Dose– response curves for tested fluorophores are shown in Figure 6. For fluorophores with strong absorp- tion in the NIR range, namely, ICG and IRDye800, mini- mal changes in cell viability were observed at 2× clinical imaging concentration ranges and illumination parame- ters. IRDye800 products showed no significant decrease in cell viability at 1 and 6 J/cm2. For ICG, a modest decrease in cell viability (to 86%) was observed at 30 μM at 6 J/cm2. The cell viability plateaus to 39% in the 100– 175 μM ex- tended concentration range at 6 J/cm2, while no significant effect in cell viability is observed at 1 J/cm2. MB products at 1 J/cm2 did not affect cell viability compared to NO IRR at the 2× clinical imaging concentration range. However, upon increasing the He to 6 J/cm2, significant photocyto- toxicity was produced at concentrations between 5 and 30 μM. No significant difference between the NO IRR and IRR response curves was observed at concentrations above 60 μM. Proflavine products showed a significant de- crease in cell viability at He as low as 0.6 J/cm2. The dose– response curve for proflavine appeared similar to those of positive controls and known RMS generators, BPD and RB. Proflavine at the 2× clinical imaging concentration range showed no change in cell viability upon increasing the concentration in the NO IRR group. However, a de- crease in cell viability by almost 100% was observed for proflavine in the NO IRR group at higher concentrations (50– 100 μM). In the case of MB and proflavine, significant inherent toxicity of the compounds was observed beside the irradiation effect in the expanded OECD concentra- tion ranges, highlighting the light- dependent and light- independent cytotoxicity of the agents. MPE prediction model and data analysis A tabular summary of the calculated MPE values for all the fluorophores at tested concentration ranges (2× clini- cal imaging concentration ranges and OECD extended concentration ranges) and illumination parameters is shown in Tables 3 and 4, respectively. Table 3 also com- pares the MPE values with the singlet oxygen production factor calculated at the 2× clinical imaging concentration ranges obtained from our previous cell- free study.56 These results demonstrate the dependence of photocytotoxicity prediction on fluorophore type, concentration, and He. The dose– response curves for each fluorophore tested He was input into the phototox v.2 software to obtain MPE prediction model values using Equation (1). The results obtained from our previous study on the singlet- oxygen production by these agents showed substantial levels of singlet- oxygen generation by RB and MB.56 Other fluorophores— ICG, IRDye800, and proflavine— showed minimal singlet oxygen generation, while BPD was not tested previously. The fluorophore products can be catego- rized into three groups: phototoxic, non- phototoxic, and equivocal phototoxic, based on the MPE prediction model values (Equation  1) calculated from the dose– response curve summarized in Tables 3 and 4. Fluorophore prod- ucts generated by known RMS generators— RB and BPD— were considered phototoxic at 2× clinical imaging and OECD extended concentration ranges as expected. IRDye800 products at 2× clinical imaging concentrations exhibited low photocytotoxicity at He 1 J/cm2. IRDye800 illuminated at 6 J/cm2 did not show any significant dif- ference from the products generated at 1 J/cm2 and thus was predicted to be non- phototoxic at all tested He. MB products exhibited no photocytotoxicity at J/cm2 in the 2× clinical imaging concentration ranges. However, sig- nificant toxicity by MB was observed in the NO IRR group at extended OECD concentrations. An overlap in the NO IRR and IRR dose– response curves of MB at an extended OECD concentration range at 1 J/cm2 was observed due to minimal IRR- based toxicity at the tested He. Hence, MB was predicted to be non- phototoxic due to minimal photo- toxic effects in spite of significant inherent toxicity at the tested concentration range. At higher He, the MPE value depicts MB products as phototoxic or non- phototoxic at 2× clinical imaging and OECD extended concentration ranges, respectively. ICG at 2× clinical imaging concen- tration range was predicted to be non- phototoxic at 1 J/ cm2. However, ICG products at an extended OECD con- centration range were predicted to be equivocal photo- toxic. At He of 6 J/cm2, ICG products were predicted to be phototoxic irrespective of the tested concentration range. Proflavine products were predicted to be phototoxic at He as low as 0.6 J/cm2, irrespective of the concentration range. DISCUSSION Light- activated generation of RMS by a fluorescent agent is governed by various factors, such as excitation wave- length, He, irradiance, and the concentration of the con- trast agent.57 In this study, the goal was to identify optimal procedures and evaluate the utility of the test method for assessing the photocytotoxic potential of clinically used fluorescence imaging products using the NRU assay as in- dicated in OECD 432. Our results with the modified 3T3 NRU assay exhibited expected trends and predictions of photocytotoxicity for known phototoxic agents BPD and RB. BPD and RB showed a concentration- dependent de- crease in cell viability at He as low as 0.2 J/cm2. RB is a 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 12 | PHOTOCHEMISTRY AND PHOTOBIOLOGY 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 13VIG et al. known singlet oxygen generator and has been reported to be used as an antimicrobial PDT agent.58 BPD has been known to produce large quantities of RMS and is used as a photodynamic agent for treating age- related macu- lar degeneration.59 It is also worth noting that for agents like RB and BPD, with higher phototoxic potential, He as low as 0.2 J/cm2 produced MPE prediction values in the phototoxic range irrespective of the 2× clinical imaging concentration ranges and OECD extended concentra- tion ranges. Given that significantly reduced cell viability is observed with BPD and RB at low light doses within the range of clinically used concentrations, these agents are undesirable for imaging purposes as phototoxicity is likely. F I G U R E 6 Dose– response curves of test fluorophores from the NRU assay. Experimental results for (A and B) proflavine; (C and D) MB; (E and F) ICG; and (G and H) IRDye800 at clinically relevant concentrations (on the left) and irradiation wavelengths, as compared with OECD432- recommended testing concentration ranges of up to 100 μg/mL (on the right) for solar- radiation assessments. The shaded region in the plots is a comparative representation of the clinically relevant 2× dose range of the test agent within the OECD- recommended concentration range. IRR, irradiation; NRU, neutral red uptake. T A B L E 3 Comparison of MPE prediction model values of singlet- oxygen (SO) production factor from the previous cell- free study calculated at clinical (2×) concentration range. Fluorophore Radiant exposure (J/ cm2) Conc. range- clinical (2×) (μM) Excitation wavelength (nm) MPE at clinical conc. range Phototoxic prediction at 2× clinical conc. range SO production factor [(F/He)/μM] at 2× clinical conc. range (μM) RB 0.2 0.5– 3 520 0.344 Phototoxic 0.70 ± 0.001 BPD 0.2 0.5– 2 685 0.701 Phototoxic N/T (not tested) ICG 1 1– 30 785 0.075 Non- phototoxic 0.0012 ± 0.0002 6 1– 30 785 0.207 Phototoxic MB 1 1– 7.5 660 0.041 Non- phototoxic 0.25 ± 0.01 6 1– 7.5 660 0.222 Phototoxic Proflavine 0.6 1– 12 450 0.262 Phototoxic 0.0080 ± 0.0014 IRDye800 1 1– 20 785 0.003 Non- phototoxic 0.0021 ± 0.00024 6 1– 20 785 0.005 Non- phototoxic Note: Interpretation of calculated MPE values based on the OECD TG 432: MPE <0.1 (No phototoxicity); MPE ≥0.1 and <0.15 (Equivocal phototoxicity); MPE ≥0.15 (Phototoxicity). Abbreviations: BPD, benzoporphyrin derivative derivative; ICG, indocyanine green; MB, methylene blue; MPE, mean photo effect; RB, rose bengal. T A B L E 4 Tabular summary of the MPE prediction model values calculated at the OECD extended concentration ranges with optimized illumination parameter combinations for fluorophore products, using Phototox V.2. software. Fluorophore Radiant exposure (J/cm2) Conc. range- OECD (μM) Excitation wavelength (nm) MPE at OECD conc. range Phototoxic prediction at OECD conc. range RB 0.2 0.5– 10 520 0.424 Phototoxic BPD 0.2 0.125– 10 685 0.757 Phototoxic ICG 1 1– 175 785 0.113 Equivocal phototoxic 6 1– 175 785 0.538 Phototoxic MB 1 1– 500 660 0.079 Non- phototoxic 6 1– 500 660 0.305 Phototoxic Proflavine 0.6 1– 1000 450 0.294 Phototoxic IRDye800 1 1– 75 785 0.016 Non- phototoxic 6 1– 75 785 0.018 Non- phototoxic Note: Interpretation of calculated MPE values based on the OECD TG 432: MPE <0.1 (No phototoxicity); MPE ≥0.1 and <0.15 (Equivocal phototoxicity); MPE ≥0.15 (Phototoxicity). Abbreviations: BPD, benzoporphyrin derivative derivative; ICG, indocyanine green; MB, methylene blue; MPE, mean photo effect; RB, rose bengal. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 14 | PHOTOCHEMISTRY AND PHOTOBIOLOGY The dose– response curve of proflavine showed a trend similar to that of positive controls, BPD and RB. This trend was different from the results gathered in our previ- ous study, where proflavine exhibited low singlet oxygen generation (Table 3).56 The literature indicates that dam- age to cellular proteins by proflavine primarily involves superoxide ions and hydroxyl radicals.59– 62 Proflavine is a DNA- intercalating agent and accumulates in the nuclei. The overall photocytotoxicity observed in our in vitro test method could be due to the production of diverse RMS (singlet oxygen, hydroxyl radicals, and superoxide ions) generated by proflavine upon light activation. This appar- ent discrepancy between the results of our current and prior studies further highlights the need for phototoxicity screening with assays that can identify a range of mecha- nisms for photocytotoxic effects. MB showed no significant change in cell viability in the 2× clinical concentration range at 1 J/cm2. However, cell viability was reduced to 50% in the presence of 6 J/ cm2 at 4.5 μM. The decrease in cell viability upon photo- activation can be correlated with a high amount of singlet oxygen generation at He of 6 J/cm2 by MB reported in our previous study using a cell- free assay at the test illumina- tion conditions (Table 3). However, the amount of singlet oxygen generated at 1 J/cm2 was insufficient to cause a significant change in cell viability at the 2× clinical imag- ing concentration range. MB is also a known DNA inter- calating agent63 and has been shown to cause increased DNA damage in Barrett's esophagus after light activation during chromoendoscopy.64 MB cytotoxicity at 30 μM was seen with and without irradiation. Thus, a concentration- dependent decrease in cell viability in the extended OECD concentration range for MB is likely a consequence of both phototoxicity and the inherent toxicity of the agent. Despite the clinical history of MB, our results and litera- ture indicate that MB can generate phototoxicity due to potential RMS generation65 and cause phototoxicity fol- lowing MB- assisted imaging procedures.66– 68 Hence, it is critical to identify safe fluorophore concentrations and ex- posure levels for MB to be used safely for diagnostic imag- ing. Thus, prior literature appears to support the need for a phototoxicity testing regime that accounts for specific light and contrast agent dose levels. IRDye800 showed no significant effect on cell viabil- ity at the tested concentration and illumination parame- ters in the study. The results obtained for IRDye800 with the modified approach are consistent with clinical trials demonstrating no adverse effects with the agent during clinical imaging procedures.15,45 IRDye800 and ICG are mainly used as imaging agents and showed the lowest sin- glet oxygen generation potential at the tested 2× clinical imaging concentration range in our previous study. ICG is a commonly used dye for retinal angiography, evaluation of cardiac output, and liver function, with a history of safety after intravenous administration.69,70 ICG has low singlet oxygen quantum yield (0.077) and also showed low singlet oxygen generation potential in our previous study.56 ICG dose– response curves at 1 and 6 J/cm2 did not significantly change cell viability at 2× clinical con- centrations. The cell viability plateaus at high testing concentrations (90– 175 μM) as a function of He. A simi- lar trend was observed with MCF- 7 and HepG2 cell lines upon photoactivation of ICG in a prior study.71 A previous study has shown that an increase in ICG concentration in an aqueous solution causes a decrease in the fluorescence intensity of the agent due to self- quenching effects.72 Although ICG is FDA- approved as a contrast agent, it has also been explored in the literature as a photosensitizer at high light doses (50 mW/cm2, 785 nm LED light, 30 min) and concentrations.73 Photosensitization of ICG produces singlet oxygen that rapidly oxidizes the agent leading to its decomposition into carbonyl compounds generated by cycloaddition of singlet oxygen.74 ICG is also an effective light absorber for photothermal therapy applications.75,76 It has been previously reported by Holzer et al. that most of the light energy absorbed by ICG is converted to heat by internal conversion.77 Thus, the photocytotoxic effect of ICG observed in our study could be due to a combination of RMS generation and photothermal effects. Additional support for our results from the modified NRU assay ap- proach comes from the MPE prediction model values cal- culated using the Phototox software and the main use of these fluorophores. The MPE prediction model is based on a comparison of the complete concentration- response curves obtained at the tested concentrations and He com- binations. An MPE can be calculated within a concentra- tion range where at least one of the dose– response curves (NO IRR or IRR) still exhibits at least 10% response. The methods implemented in this study deviate from standard approaches in that they focus on generating re- sults for clinically relevant light and agent doses. This ap- proach may enable the same contrast agent to be used for imaging and phototherapy, depending on concentration and illumination. Additionally, evaluating overall photo- cytotoxicity produced using the in vitro 3T3 NRU assay may help determine a safe concentration and illumination dose range for clinical imaging. The proposed methods will facilitate the development of innovative diagnostic technologies that are safe for clinical use and benefit pub- lic health. Results from this study also represent critical scientific information that may provide direction toward establishing consensus standards for contrast- enhanced fluorescent imaging products. Although there is generally good agreement between our experimental results and the literature, there are some issues with the NRU approach. The premise underlying 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense | 15VIG et al. this assay is that a cytotoxic chemical interferes with nor- mal lysosomal uptake, representing the number of viable cells. Intact lysosomal integrity is only maintained in viable cells.78 The degree of cell viability inhibition and associated phototoxicity for an agent would be affected by the concen- tration of the test agent used, inherent cytotoxicity of the agent without irradiation, and the RMS generated at given illumination parameters. The solubility of the test agent in the solvent (Table S1) and cell media should also be taken into consideration since hydrophobic contrast agents can aggregate at higher concentrations which might affect the cell viability upon photoactivation. Additionally, any test agent having a localized effect on the lysosomes will result in low cell number and reduced viability. The induction of precipitation of the neutral dye by some chemical agents might also impact the accuracy of the absorbance readings used to detect NR uptake.79,80 The current modified in vitro approach, although robust in capturing the overall effect of photoactivation of a fluo- rophore, is not informative in determining the mechanism of action of phototoxicity. Our previous study used a cell- free assay to quantify singlet oxygen produced by contrast- enhanced fluorescence imaging agents.56 This produced essential information about a potential phototoxicity mech- anism of interest for the tested fluorophores. Together, the cell- free and the in vitro phototoxicity assessment strategies could constitute an informative test methodology to evalu- ate the potential phototoxicity of emerging fluorescence im- aging products. Based on our current and previous results, the phototoxicity testing of fluorescence imaging products should include initial testing with the cell- free and the in vitro phototoxicity assessment strategy. While the modi- fied NRU assay would help evaluate the biological effects caused by the test fluorophore products, the cell- free assay will provide a better understanding of the RMS involved in associated phototoxicity upon photoactivation. Our future work will focus on (1) developing test methods that can be standardized to quantify other RMS (hydroxyl radicals, su- peroxide ions, etc.) generated by fluorescent contrast agents and (2) in vivo test methods to detect RMS generation and associated photocytotoxicity of fluorescence contrast agents used for diagnostic purposes. CONCLUSION In this study, we have proposed and evaluated a modi- fied in vitro method based on the neutral red uptake assay (NRU) for assessing the photocytotoxicity of fluorescence products generated under clinical conditions. Results ob- tained from this study align well with our recent work on reactive oxygen species generation with a cell- free assay and prior published results on in vivo phototoxic effects of fluorophores. Therefore, this approach may represent a practical and least burdensome approach for phototoxic- ity screening of emerging visible to near- infrared fluores- cence imaging products. ACKNOWLEDGMENTS This work is supported by the NSF- FDA scholar- in- residence Program CBET- 2037815. Ms. Shruti Vig is sup- ported by the MPower Fellowship. Mr. Brandon Gaitan is supported by the Clark Doctoral Fellowship. Mr. Lucas Frankle is supported by the University of Maryland ASPIRE Fellowship. DISCLAIMER The mention of commercial products, their sources, or their use in connection with material reported herein is not to be construed as an actual or implied endorsement of such products by the Department of Health and Human Services. ORCID Shruti Vig  https://orcid.org/0000-0002-0534-1571 Huang- Chiao Huang  https://orcid. org/0000-0002-5406-0733 REFERENCES 1. Tipirneni KE, Rosenthal EL, Moore LS, et al. Fluorescence im- aging for cancer screening and surveillance. Mol Imaging Biol. 2017;19(5):645- 655. 2. Stummer W, Suero Molina E. Fluorescence imaging/agents in tumor resection. Neurosurg Clin N Am. 2017;28(4):569- 583. 3. Pierce MC, Javier DJ, Richards- Kortum R. Optical contrast agents and imaging systems for detection and diagnosis of can- cer. Int J Cancer. 2008;123(9):1979- 1990. 4. Yamamoto M. Indocyanine green angiography for intra- operative assessment in vascular surgery. Eur J Vasc Endovasc Surg. 2012;43(4):426- 432. 5. Brancato R, Bandello F, Lattanzio R. Iris fluorescein angiogra- phy in clinical practice. Surv Ophthalmol. 1997;42(1):41- 70. 6. Vahrmeijer AL. Image- guided cancer surgery using near- infrared fluorescence. Nat Rev Clin Oncol. 2013;10(9):507- 518. 7. Palmieri G, Cofano F, Salvati LF, et al. Fluorescence- guided sur- gery for high- grade gliomas: state of the art and new perspec- tives. Technol Cancer Res Treat. 2021;20:1- 11. 8. Polom W, Markuszewski M, Rho YS, Matuszewski M. Use of invisible near- infrared light fluorescence with indocyanine green and methylene blue in urology. Part 2. Cent Eur J Urol. 2014;67(3):310- 313. 9. van Manen L, Handgraaf HJM, Diana M, et al. A practical guide for the use of indocyanine green and methylene blue in fluorescence- guided abdominal surgery. J Surg Oncol. 2018;118(2):283- 300. 10. Tummers QR, Verbeek FP, Schaafsma BE, et al. Real- time in- traoperative detection of breast cancer using near- infrared fluorescence imaging and methylene blue. Eur J Surg Oncol. 2014;40(7):850- 858. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://orcid.org/0000-0002-0534-1571 https://orcid.org/0000-0002-0534-1571 https://orcid.org/0000-0002-5406-0733 https://orcid.org/0000-0002-5406-0733 https://orcid.org/0000-0002-5406-0733 16 | PHOTOCHEMISTRY AND PHOTOBIOLOGY 11. van der Vorst JR, Schaafsma BE, Verbeek FP, et al. Intraoperative near- infrared fluorescence imaging of parathy- roid adenomas with use of low- dose methylene blue. Head Neck. 2014;36(6):853- 858. 12. Cwalinski T, Polom W, Marano L, et al. Methylene blue— current knowledge, fluorescent properties, and its future use. J Clin Med. 2020;9(11):35- 38. 13. Varghese P, Abdel- Rahman AT, Akberali S, Mostafa A, Gattuso JM, Carpenter R. Methylene blue dye— a safe and effective alternative for sentinel lymph node localization. Breast J. 2008;14(1):61- 67. 14. Kawedia JD, Zhang YP, Myers AL, et al. Physical and chemical stability of proflavine contrast agent solutions for early detec- tion of oral cancer. J Oncol Pharm Pract. 2016;22(1):21- 25. 15. Rosenthal EL, Warram JM, de Boer E, et al. Safety and tumor specificity of cetuximab- IRDye800 for surgical navigation in head and neck cancer. Clin Cancer Res. 2015;21(16):3658- 3666. 16. Hernot S, van Manen L, Debie P, Mieog JSD, Vahrmeijer AL. Latest developments in molecular tracers for fluorescence image- guided cancer surgery. Lancet Oncol. 2019;20(7):e354 - e367. 17. Feasibility trial for real- time identification of the ureters with a methylene- blue. NCT05111808. November 8, 2021. Accessed April 24, 2023. https://clini caltr ials.gov/show/ NCT05 111808 18. Fluorescent imaging and methylene blue: ureter study. NCT03177070. June 6, 2017. https://clini caltr ials.gov/show/ NCT03 177070 19. Intra- operative infra- red fluorescent imaging in thyroid and parathyroid surgery. NCT02089542. March 17, 2014. Accessed April 24, 2023. https://clini caltr ials.gov/show/NCT02 089542 20. Near infrared fluorescent imaging in thyroid and parathy- roid surgery with the Fluobeam (TM) system of Fluoptics. NCT01598727. May 15, 2012. Accessed April 24, 2023. https:// clini caltr ials.gov/ct2/show/NCT01 598727 21. COVID- 19 treatment using methylene blue and photodynamic therapy. NCT04933864. June 22, 2021. Accessed June 16, 2023. https://clini caltr ials.gov/show/NCT04 933864 22. Efficacy of methylene blue mediated photodynamic therapy for primary localized cutaneous amyloidosis treatment in Asians. NCT03068208. March 1, 2017. Accessed June 16, 2023. https:// clini caltr ials.gov/show/NCT03 068208 23. Comparison of efficacy and safety between methylene blue- mediated photodynamic therapy and 5% amorolfine nail lac- quer for toenail onychomycosis treatment. NCT03098342. March 31, 2017. Accessed June 16, 2023. https://clini caltr ials. gov/show/NCT03 098342 24. King Saud U. ICG- PDT, periimplantitis, diabetes mellitus. NCT04833569. April 6, 2021. Accessed June 16, 2023. https:// clini caltr ials.gov/show/NCT04 833569 25. Single session of antimicrobial photodynamic therapy using indocyanine green. NCT02043340. January 23, 2014 (CTG Labs— NCBI). 26. Center for Devices and Radiological Health, FDA. Laser products— conformance with IEC 60825– 1 Ed. 3 and IEC 60601– 2- 22 Ed. 3.1 (Laser Notice No. 56). 2019. Accessed December 22, 2022. https://www.fda.gov/media/ 11012 0/download 27. International Electrotechnical Commission (IEC). Photobiological safety of lamps and lamp systems, IEC 62471:Ed. 1.0: 2006. 2006– 2007. 2006. Accessed January 9, 2023. https://webst ore.iec.ch/publi catio n/7076 28. Khandpur S, Porter RM, Boulton SJ, Anstey A. Drug- induced photosensitivity: new insights into pathomechanisms and clin- ical variation through basic and applied science. Br J Dermatol. 2017;176(4):902- 909. 29. Kim K, Park H, Lim KM. Phototoxicity: its mechanism and an- imal alternative test methods. Toxicol Res. 2015;31(2):97- 104. 30. Baptista MS, Cadet J, Di Mascio P, et al. Type I and type II pho- tosensitized oxidation reactions: guidelines and mechanistic pathways. Photochem Photobiol. 2017;93(4):912- 919. 31. Borenfreund E, Puerner JA. Toxicity determined in vitro by morphological alterations and neutral red absorption. Toxicol Lett. 1985;24(2):119- 124. 32. Bulychev A, Trouet A, Tulkens P. Uptake and intracellular dis- tribution of neutral red in cultured fibroblasts. Exp Cell Res. 1978;115(2):343- 355. 33. Thorne D, Wieczorek R, Fukushima T. A survey of aerosol expo- sure systems relative to the analysis of cytotoxicity: a Cooperation Centre for Scientific Research Relative to Tobacco (CORESTA) perspective. Toxicol Res Appl. 2021;5:239784732110222. 34. Borenfreund E, Puerner JA. A simple quantitative procedure using monolayer cultures for cytotoxicity assays (HTD/NR- 90). J Tissue Cult Methods. 1985;9(1):7- 9. 35. Lynch AM, Wilcox P. Review of the performance of the 3T3 NRU in vitro phototoxicity assay in the pharmaceutical indus- try. Exp Toxicol Pathol. 2011;63(3):209- 214. 36. Ceridono M, Tellner P, Bauer D, et al. The 3T3 neutral red up- take phototoxicity test: practical experience and implications for phototoxicity testing— the report of an ECVAM- EFPIA workshop. Regul Toxicol Pharmacol. 2012;63(3):480- 488. 37. OECD. Test No. 432: In Vitro 3T3 NRU Phototoxicity Test, OECD Guidelines for the Testing of Chemicals, Section 4. OECD Publishing; 2019. doi:10.1787/9789264071162- en 38. Prieto Peraita M, Griesinger C, Amcoff S, Whelan M. EURL ECVAM Recommendation on the 3T3 Neutral Red Uptake Cytotoxicity Assay for Acute Oral Toxicity Testing. Publications Office of the European Union; 2013 (JRC79556). 39. FDA. S10 photosafety evaluation of pharmaceuticals. 2015. Accessed January 9, 2023. https://www.fda.gov/media/ 85076/ download 40. Verbeek FPR, van der Vorst JR, Schaafsma BE, et al. Intraoperative near infrared fluorescence guided identification of the ureters using low dose methylene blue: a first in human experience. J Urol. 2013;190(2):574- 579. 41. Tagaya N, Yamazaki R, Nakagawa A, et al. Intraoperative identification of sentinel lymph nodes by near- infrared flu- orescence imaging in patients with breast cancer. Am J Surg. 2008;195(6):850- 853. 42. Zhu B, Sevick- Muraca EM. A review of the performance of near- infrared fluorescence imaging devices used in clinical studies. Br J Radiol. 2015;88(1045):20140547. 43. Di Stefano AFD, Radicioni MM, Vaccani A, et al. Methylene blue MMX® tablets for chromoendoscopy. Bioavailability, colon staining and safety in healthy volunteers undergoing a full colo- noscopy. Contemp Clin Trials. 2018;71:96- 102. 44. Matsui A, Tanaka E, Choi HS, et al. Real- time, near- infrared, fluorescence- guided identification of the ureters using methy- lene blue. Surgery. 2010;148(1):78- 86. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://clinicaltrials.gov/show/NCT05111808 https://clinicaltrials.gov/show/NCT05111808 https://clinicaltrials.gov/show/NCT03177070 https://clinicaltrials.gov/show/NCT03177070 https://clinicaltrials.gov/show/NCT02089542 https://clinicaltrials.gov/ct2/show/NCT01598727 https://clinicaltrials.gov/ct2/show/NCT01598727 https://clinicaltrials.gov/show/NCT04933864 https://clinicaltrials.gov/show/NCT03068208 https://clinicaltrials.gov/show/NCT03068208 https://clinicaltrials.gov/show/NCT03098342 https://clinicaltrials.gov/show/NCT03098342 https://clinicaltrials.gov/show/NCT04833569 https://clinicaltrials.gov/show/NCT04833569 https://www.fda.gov/media/110120/download https://webstore.iec.ch/publication/7076 https://doi.org//10.1787/9789264071162-en https://www.fda.gov/media/85076/download https://www.fda.gov/media/85076/download | 17VIG et al. 45. Miller SE, Tummers WS, Teraphongphom N, et al. First- in- human intraoperative near- infrared fluorescence imaging of glioblastoma using cetuximab- IRDye800. J Neurooncol. 2018;139(1):135- 143. 46. Dsouza AV, Lin H, Henderson ER, Samkoe KS, Pogue BW. Review of fluorescence- guided surgery systems: identification of key performance capabilities beyond indocyanine green im- aging. J Biomed Opt. 2016;21(8):080901. 47. Tang Y, Carns J, Polydorides AD, Anandasabapathy S, Richards- Kortum RR. In vivo white light and contrast- enhanced vital- dye fluorescence imaging of Barrett's- related neoplasia in a single- endoscopic insertion. J Biomed Opt. 2016;21(8):086004. 48. Shin D, Pierce MC, Gillenwater AM, Williams MD, Richards- Kortum RR. A fiber- optic fluorescence microscope using a consumer- grade digital camera for in vivo cellular imaging. PLoS ONE. 2010;5(6):e11218. 49. Pérez- Laguna V, García- Luque I, Ballesta S, et al. Antimicrobial photodynamic activity of rose bengal, alone or in combination with Gentamicin, against planktonic and biofilm Staphylococcus aureus. Photodiagnosis Photodyn Ther. 2018;21:211- 216. 50. Fingar VH, Kik PK, Haydon PS, et al. Analysis of acute vascular damage after photodynamic therapy using benzoporphyrin de- rivative (BPD). Br J Cancer. 1999;79(11):1702- 1708. 51. Tsoukas MM, Lin GC, Lee MS, Anderson RR, Kollias N. Predictive dosimetry for threshold phototoxicity in photody- namic therapy on normal skin: red wavelengths produce more extensive damage than blue at equal threshold doses. J Invest Dermatol. 1997;108(4):501- 505. 52. Repetto G, del Peso A, Zurita JL. Neutral red uptake assay for the estimation of cell viability/cytotoxicity. Nat Protoc. 2008;3(7):1125- 1131. 53. Holzhütter H- G. A general measure of in vitro phototoxicity derived from pairs of dose– response curves and its use for pre- dicting the in vivo phototoxicity of chemicals. Altern Lab Anim. 1997;25(4):445- 462. 54. Spielmann H, Balls M, Dupuis J, et al. The international EU/ COLIPA In vitro phototoxicity validation study: results of phase II (blind trial). Part 1: the 3T3 NRU phototoxicity test. Toxicol In Vitro. 1998;12(3):305- 327. 55. Goulart Rde C, Bolean M, Tde Paulino P, et al. Photodynamic therapy in planktonic and biofilm cultures of Aggregatibacter actinomycetemcomitans. Photomed Laser Surg. 2010;28(Suppl 1):S53- S60. 56. Gaitan B, Frankle L, Vig S, et al. Quantifying the photochemi- cal damage potential of contrast- enhanced fluorescence imag- ing products: singlet oxygen production. Photochem Photobiol. 2022;98(4):736- 747. 57. De Silva P, Saad MA, Thomsen H, Bano S, Ashraf S, Hasan T. Photodynamic therapy, priming and optical imaging: poten- tial co- conspirators in treatment design and optimization— a Thomas Dougherty Award for Excellence in PDT paper. J Porphyr Phthalocyanines. 2020;24:1320- 1360. 58. Arboleda A, Miller D, Cabot F, et al. Assessment of rose bengal versus riboflavin photodynamic therapy for inhibition of fungal keratitis isolates. Am J Ophthalmol. 2014;158(1):64- 70.e2. 59. Kramer M, Miller JW, Michaud N, et al. Liposomal benzopor- phyrin derivative verteporfin photodynamic therapy: selec- tive treatment of choroidal neovascularization in monkeys. Ophthalmology. 1996;103(3):427- 438. 60. Gatasheh MK, Malik A, Ola MS, Alhomida AS. Photo- activated proflavine degrades protein and impairs enzyme activity: in- volvement of hydroxyl radicals. Toxicol Rep. 2022;9:78- 86. 61. Piette J, Decuyper J, Machiroux R, Calberg- Bacq CM, van de Vorst A, Lion Y. Visible- light- induced OH radicals in DNA- proflavine complexes: an e.p.r. and spin trapping study. Int J Radiat Biol Relat Stud Phys Chem Med. 1982;42(2):151- 161. 62. Gatasheh MKM, Subbaram K, Kannan H, Naseem I. Generation of oxygen free radicals by proflavine: implication in protein degradation. HAYATI J Biosci. 2017;24(3):118- 123. 63. Bradley DF, Stellwagen NC, O'Konski CT, Paulson CM. Electric birefringence and dichroism of acridine orange and meth- ylene blue complexes with polynucleotides. Biopolymers. 1972;11(3):645- 652. 64. Olliver JR, Wild CP, Sahay P, Dexter S, Hardie LJ. Chromoendoscopy with methylene blue and associated DNA damage in Barrett's esophagus. Lancet. 2003;362(9381):373- 374. 65. Epe B, Hegler J, Wild D. Singlet oxygen as an ultimately re- active species in Salmonella typhimurium DNA damage induced by methylene blue/visible light. Carcinogenesis. 1989;10(11):2019- 2024. 66. Kluk J, Charles- Holmes R, Markham D. Acute phototoxic- ity induced by methylene blue- assisted parathyroidectomy: an unreported risk of a common procedure. Br J Dermatol. 2012;166(4):907- 908. 67. Lambrecht P, Vandeplas G, van Slycke S, Smet PF, Brusselaers N, Vermeersch H. Phototoxic reaction after parathyroid sur- gery: case report and review of the literature. Acta Clin Belg. 2012;67(6):438- 441. 68. Maguire CA, Sharma A, Alarcon L, et al. Histological features of methylene blue- induced phototoxicity administered in the con- text of parathyroid surgery. Am J Dermatopathol. 2017;39(8):e1 10- e115. 69. Alander JT, Kaartinen I, Laakso A, et al. A review of indo- cyanine green fluorescent imaging in surgery. Int J Biomed Imaging. 2012;2012:940585. 70. Boni L, David G, Mangano A, et al. Clinical applications of in- docyanine green (ICG) enhanced fluorescence in laparoscopic surgery. Surg Endosc. 2015;29(7):2046- 2055. 71. El- Daly SM, Gamal- Eldeen AM, Abo- Zeid MA, Borai IH, Wafay HA, Abdel- Ghaffar AR. Photodynamic therapeutic activity of indocyanine green entrapped in polymeric nanoparticles. Photodiagnosis Photodyn Ther. 2013;10(2):173- 185. 72. Saxena V, Sadoqi M, Shao J. Degradation kinetics of indocy- anine green in aqueous solution. J Pharm Sci. 2003;92(10): 2090- 2097. 73. Lim H- J, Oh CH. Indocyanine green- based photodynamic ther- apy with 785nm light emitting diode for oral squamous cancer cells. Photodiagnosis Photodyn Ther. 2011;8(4):337- 342. 74. Engel E, Schraml R, Maisch T, et al. Light- induced decom- position of indocyanine green. Invest Ophthalmol Vis Sci. 2008;49(5):1777- 1783. 75. Shirata C, Kaneko J, Inagaki Y, et al. Near- infrared photother- mal/photodynamic therapy with indocyanine green induces apoptosis of hepatocellular carcinoma cells through oxidative stress. Sci Rep. 2017;7(1):13958. 76. Zheng X, Zhou F, Wu B, Chen WR, Xing D. Enhanced tumor treatment using biofunctional indocyanine green- containing nanostructure by intratumoral or intravenous injection. Mol Pharm. 2012;9(3):514- 522. 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 18 | PHOTOCHEMISTRY AND PHOTOBIOLOGY 77. Holzer W, Mauerer M, Penzkofer A, et al. Photostability and thermal stability of indocyanine green. Photochem Photobiol. 1998;47(2):155- 164. 78. Wang F, Gómez- Sintes R, Boya P. Lysosomal membrane perme- abilization and cell death. Traffic. 2018;19(12):918- 931. 79. Barile FA. 3— Continuous cell lines as a model for drug tox- icity assessment. In: Castell JV, Gmez- Lechn MJ, eds. In Vitro Methods in Pharmaceutical Research. Academic Press; 1997:33- 54. 80. McGaw LJ, Elgorashi EE, Eloff JN. 8— Cytotoxicity of African Medicinal Plants Against Normal Animal and Human Cells, Toxicological Survey of African Medicinal Plants. Elsevier; 2014:181- 233. 81. Unno N, Inuzuka K, Suzuki M, et al. Preliminary experience with a novel fluorescence lymphography using indocyanine green in patients with secondary lymphedema. J Vasc Surg. 2007;45(5):1016- 1021. 82. Takeuchi M, Sugie T, Abdelazeem K, et al. Lymphatic mapping with fluorescence navigation using indocyanine green and ax- illary surgery in patients with primary breast cancer. Breast J. 2012;18(6):535- 541. 83. Imai T, Takahashi K, Goto F, Morishita Y. Measurement of blood concentration of indocyanine green by pulse dye densitometry— comparison with the conventional spectropho- tometric method. J Clin Monit Comput. 1998;14(7– 8):477- 484. 84. Gao RW, Teraphongphom N, de Boer E, et al. Safety of panitumumab- IRDye800CW and cetuximab- IRDye800CW for fluorescence- guided surgical navigation in head and neck can- cers. Theranostics. 2018;8(9):2488- 2495. 85. Wide- field and high- resolution in vivo imaging in visualizing lesions in patients with oral neoplasia undergoing surgery. NCT012699190. December 30, 2010. Accessed April 24, 2023. https://clini caltr ials.gov/show/NCT01 269190 SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article: Vig S, Gaitan B, Frankle L, et al. Test method for evaluating the photocytotoxic potential of fluorescence imaging products. Photochem Photobiol. 2023;00:1-18. doi:10.1111/php.13836 17511097, 0, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/php.13836 by U niversity O f M aryland, W iley O nline L ibrary on [18/06/2024]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://clinicaltrials.gov/show/NCT01269190 https://doi.org/10.1111/php.13836 Test method for evaluating the photocytotoxic potential of fluorescence imaging products Abstract INTRODUCTION MATERIALS AND METHODS Overview Cell culture Fluorophore concentrations and illumination levels Optical exposure system setup and validation Modified neutral red uptake (NRU) assay Data analysis Statistical analysis RESULTS Validation of illumination system and NRU assay Evaluation of the photocytotoxicity using the modified NRU assay MPE prediction model and data analysis DISCUSSION CONCLUSION ACKNOWLEDGMENTS DISCLAIMER REFERENCES