Internalization-Dependent EGFR-Targeted Photoactivation by Cetuximab-I21 Conjugates

Abstract

Photosensitizer-based antibody conjugates can provide tumor-selective phototoxicity by combining receptor-mediated targeting with local light activation. IR700-based near-infrared photoimmunotherapy represents the clinical benchmark for this strategy; however, photosensitizers with alternative cellular trafficking and phototoxic mechanisms remain of interest. We developed and characterized 2 cetuximab-I21 conjugates, cetuximab-I21-2 and cetuximab-I21-3, targeting epidermal growth factor receptor (EGFR)-expressing tumors. The chemical structures and key photophysical properties of the I21 payload-linkers are disclosed in this revision. Both conjugates achieved a drug-to-antibody ratio of 8 and demonstrated strictly light-dependent cytotoxicity in EGFR-positive cells, with IC₅₀ values of 0.051–1.093 μg/mL in A431 cells and no detectable dark toxicity. EGFR-low cells showed >400-fold reduced sensitivity, supporting receptor-dependent selectivity. Mechanistic studies revealed receptor-mediated internalization and endolysosomal trafficking, with peak lysosomal colocalization at 3 hours (Pearson r=0.81), followed by mitochondrial membrane depolarization and dose-dependent apoptosis-associated cell death. In vivo fluorescence imaging demonstrated tumor-selective accumulation with sustained retention through 96 hours. In A431 xenografts, cetuximab-I21-3 with triple irradiation (200 J/cm²×3) achieved 50.42% tumor growth inhibition (P=0.0002). These findings establish cetuximab-I21 as a light-activated, EGFR-targeted antibody-photosensitizer conjugate with a mechanism distinct from that of canonical IR700-based NIR-PIT. Direct IR700 comparison, immune activation studies, photosafety evaluation, and light-dose optimization will be required to define its translational potential.

Antibody-drug conjugates (ADCs) have emerged as one of the most promising therapeutic modalities in modern oncology, representing a targeted therapeutic modality with relevance to precision immunotherapy that combines the exquisite targeting specificity of monoclonal antibodies with the cytotoxic potency of small-molecule payloads.^1–3 The fundamental concept underlying ADC technology is elegantly simple: by covalently attaching cytotoxic agents to tumor-targeting antibodies through chemical linkers, these biopharmaceutical agents can selectively deliver their toxic payload to cancer cells while minimizing exposure to healthy tissues. This targeted approach addresses one of the central challenges in cancer chemotherapy—achieving sufficient drug concentrations at the tumor site while limiting systemic toxicity that often dose-limits conventional chemotherapeutic agents.

The clinical development of ADCs has witnessed remarkable progress since the regulatory approval of gemtuzumab ozogamicin in 2000, which represented the first ADC to receive marketing authorization from the US Food and Drug Administration.^4,5 Despite its subsequent voluntary withdrawal in 2010 due to concerns regarding hepatotoxicity and lack of clinical benefit in confirmatory trials, the ADC field has evolved substantially over the past 2 decades. The reapproval of gemtuzumab ozogamicin in 2017 with revised dosing recommendations, along with the successful development of brentuximab vedotin, ado-trastuzumab emtansine, and, more recently, enfortumab vedotin and sacituzumab govitecan, has validated the therapeutic potential of this drug class. As of 2024, over 100 ADC candidates are in clinical development across various tumor types, with 15 FDA-approved products demonstrating meaningful clinical benefit in hematological malignancies and solid tumors.^6,7

Despite these significant advances, critical pharmaceutical and clinical challenges persist that limit the broader application of ADC technology.^8,9 First, payload potency optimization remains a delicate balance—while highly potent cytotoxic agents such as auristatins and maytansinoids provide the requisite tumor cell killing, their extreme toxicity (often in the picomolar range) means that even small amounts of premature release or nonspecific uptake can cause significant off-target effects. Second, tumor penetration efficiency is often suboptimal, particularly for large antibody-based therapeutics that must navigate the hostile tumor microenvironment characterized by elevated interstitial pressure, abnormal vasculature, and dense extracellular matrix. Third, off-target toxicity arising from target antigen expression on normal tissues, Fc receptor-mediated uptake by immune cells, or linker instability in circulation continues to narrow the therapeutic window of many ADC programs.

In addition, resistance mechanisms associated with ADC therapy have emerged as a significant clinical concern.^10,11 These include downregulation or loss of target antigen expression, impaired internalization and intracellular trafficking, enhanced drug efflux through ATP-binding cassette transporters, alterations in lysosomal function affecting payload release, and mutations in tubulin or other payload targets. The multifactorial nature of ADC resistance underscores the need for innovative approaches that can circumvent these mechanisms while maintaining the fundamental advantages of antibody-mediated targeting.

Photosensitizer-based ADCs (PS-ADCs) represent an innovative drug-delivery strategy that addresses many of these limitations by incorporating an additional layer of selectivity through light-dependent activation.^12,13 In this approach, the conjugated photosensitizer remains pharmacologically inactive until exposed to light of specific wavelengths, typically in the near-infrared (NIR) region of the electromagnetic spectrum. This dual-selectivity mechanism—combining receptor-mediated targeting with light-activated toxicity—significantly expands the therapeutic window compared with conventional cytotoxic ADCs by providing spatiotemporal control over the cytotoxic effect. The photosensitizer can accumulate in tumor tissue through antibody-mediated delivery, but therapeutic activity is initiated only when the tumor site is deliberately illuminated with the appropriate wavelength of light.

The mechanism of photodynamic therapy (PDT) involves the excitation of photosensitizer molecules from their ground state to an excited singlet state upon absorption of light energy.^14,15 Through intersystem crossing, the excited singlet state can convert to a longer-lived triplet state, which subsequently transfers energy to molecular oxygen to generate reactive oxygen species (ROS), primarily singlet oxygen (¹O₂). These highly reactive species cause oxidative damage to cellular components, including lipids, proteins, and nucleic acids, leading to cell death through multiple mechanisms. Unlike conventional chemotherapeutic agents that typically act through a single molecular target, photodynamic therapy induces a multitarget oxidative assault that is inherently resistant to the resistance mechanisms that plague conventional ADCs.

Photosensitizers typically exhibit minimal dark toxicity, meaning they cause negligible cellular damage in the absence of light activation.^16,17 This property is particularly advantageous for ADC applications, as it addresses the off-target toxicity concerns that arise from premature payload release or nonspecific uptake. Even if the PS-ADC is taken up by nontarget tissues, no therapeutic effect occurs unless those tissues are deliberately illuminated. This provides an inherent safety margin that is not achievable with conventional cytotoxic payloads, which retain their toxic potential regardless of their anatomic location.

The epidermal growth factor receptor (EGFR) represents an attractive target for PS-ADC development due to its overexpression in multiple solid tumor types.^18,19 EGFR is a transmembrane receptor tyrosine kinase that plays a central role in regulating cell proliferation, survival, differentiation, and migration. Overexpression or activating mutations of EGFR have been documented in numerous malignancies, including non–small cell lung cancer (NSCLC), colorectal cancer (CRC), head and neck squamous cell carcinoma (HNSCC), pancreatic cancer, and glioblastoma. The high expression levels on tumor cells relative to normal tissues, combined with efficient receptor-mediated endocytosis upon ligand or antibody binding, make EGFR an ideal target for antibody-mediated drug delivery.

Cetuximab is an FDA-approved chimeric IgG1 monoclonal antibody that binds to the extracellular domain of EGFR with high affinity, blocking ligand binding and receptor activation.^20,21 Beyond its direct antiproliferative effects, cetuximab demonstrates well-characterized pharmacokinetics with a terminal half-life of ∼7 days, established clinical safety profiles from extensive use in colorectal cancer and head and neck cancer, and efficient receptor-mediated internalization—properties essential for effective ADC drug delivery. Upon binding to EGFR on the cell surface, cetuximab triggers receptor dimerization and internalization through clathrin-mediated endocytosis, subsequently trafficking through the endosomal-lysosomal pathway, where payload release can occur.

The IR700 photosensitizer (IRDye 700DX, a silicon-phthalocyanine derivative) has been extensively validated for photoimmunotherapy applications.^22,23 This water-soluble, NIR-absorbing photosensitizer exhibits favorable properties, including strong absorption at ∼690 nm (allowing deeper tissue penetration than visible light photosensitizers), high singlet oxygen quantum yield, excellent photostability, and minimal dark toxicity. The cetuximab-IR700 conjugate (ASP-1929, also known as cetuximab sarotalocan) has completed clinical development and received regulatory approval in Japan in September 2020 for the treatment of unresectable locally advanced or recurrent head and neck cancer, representing the first approved photoimmunotherapy agent.^24,25 Clinical studies demonstrated objective response rates of 28%–50% with a favorable safety profile characterized primarily by local infusion-related reactions and treatment site complications.

Building upon the clinical and mechanistic precedent established by IR700-based NIR-PIT, we developed and characterized 2 cetuximab-I21 conjugates, designated cetuximab-I21-2 and cetuximab-I21-3. The I21 series was developed as a mechanistically distinct photosensitizer platform compatible with cetuximab-mediated EGFR targeting, rather than as a demonstrated replacement for IR700. The chemical structures and key photophysical properties of the I21-2 and I21-3 payload-linkers, including molecular weight, absorption maxima, conjugation handle, and DAR after conjugation, are provided in Figure S2, Supplemental Digital Content 1, http://links.lww.com/JIT/B69. In the present study, we evaluated in vitro light-dependent cytotoxicity across tumor cell lines with varying EGFR expression, characterized intracellular trafficking and subcellular localization, assessed mitochondrial injury and apoptosis-associated cell death, examined in vivo biodistribution, and evaluated antitumor activity under defined light-dosing conditions. Because a direct head-to-head comparison with cetuximab-IR700 was not performed, the present study does not claim superiority over IR700-based NIR-PIT, but instead defines the properties and limitations of cetuximab-I21 as a distinct EGFR-targeted photoactivated antibody conjugate.

MATERIALS AND METHODS

Materials and Reagents

Cetuximab (Erbitux) was obtained from Merck KGaA (Darmstadt, Germany) and Eli Lilly (Indianapolis, IN) as pharmaceutical-grade material. The I21-2 and I21-3 maleimide-functionalized photosensitizer payload-linkers, and the corresponding cetuximab-I21 conjugates were synthesized and provided by Shanghai Biophy Biological Pharmaceutical Co, Ltd. The chemical structures and selected photophysical properties of I21-2 and I21-3 are provided in Figure S2, Supplemental Digital Content 1, http://links.lww.com/JIT/B69. The payload-linkers showed red-light absorption maxima of 650.25 and 655.15 nm, respectively, in 1×PBS (pH: 7.4), and both used maleimide chemistry for cysteine conjugation. Tris (2-carboxyethyl)phosphine hydrochloride (TCEP-HCl) and other conjugation reagents were purchased from Sigma-Aldrich (St. Louis, MO). For cell-based assays, the JC-1 mitochondrial membrane potential assay kit was purchased from Beyotime Biotechnology (Shanghai, China). The Annexin V-FITC/PI apoptosis detection kit was obtained from BD Biosciences (San Jose, CA). LysoTracker Green DND-26, MitoTracker Red CMXRos, and Hoechst 33342 were purchased from Thermo Fisher Scientific (Waltham, MA). Cell Counting Kit-8 (CCK-8) was obtained from Dojindo Molecular Technologies (Kumamoto, Japan).

Cell Lines and Culture Conditions

Human cancer cell lines representing a spectrum of EGFR expression levels were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and authenticated by short tandem repeat (STR) profiling before use. A431 human epidermoid carcinoma cells were selected as the primary model system due to their exceptionally high EGFR expression (∼2×10⁶ receptors per cell). HCC-827 human lung adenocarcinoma cells harbor an EGFR exon 19 deletion mutation and were included to evaluate activity in EGFR-mutant cancer. NCI-N87 human gastric carcinoma cells express moderate levels of EGFR. Huh7 human hepatocellular carcinoma cells and A549 human lung adenocarcinoma cells express low to negligible EGFR levels and served as negative controls for receptor-dependent selectivity. Cells were cultured in appropriate media supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin at 37 °C in 5% CO₂.

Preparation and Characterization of Photoimmunotherapy Conjugates

Cetuximab-I21 conjugates were synthesized using cysteine conjugation chemistry. Briefly, cetuximab in PBS (pH: 7.4) was partially reduced to expose interchain cysteine residues and then reacted with maleimide-functionalized I21-2 or I21-3 payload-linkers to generate cetuximab-I21-2 and cetuximab-I21-3, respectively. Conjugates were purified to remove unconjugated payload-linkers. The drug-to-antibody ratio (DAR), a parameter influencing photosensitizer loading and photoactivation potency, was determined by analytical characterization of the conjugated products. The final DAR for both cetuximab-I21-2 and cetuximab-I21-3 was 8. The calculated molecular weights of cetuximab-I21-2 and cetuximab-I21-3 were ∼161.9 and 162.2 kDa, respectively. The absorption maximum of the conjugates was 652.70 and 655.15 nm in 1×PBS (pH: 7.4), respectively. The extinction coefficient and the singlet oxygen quantum yield were not determined in the present study. EGFR binding affinity was verified by flow cytometry on A431 cells.

In Vitro Cytotoxicity Assays

Cells were seeded in 96-well plates at 5000–8000 cells/well and allowed to adhere overnight. Cetuximab-I21 conjugates, or controls, were added at various concentrations and incubated for 4–24 hours to allow receptor-mediated internalization. Plates were then irradiated with 660 nm LED light at specified doses (0, 1, 5, 10, 50 J/cm²). Dark control groups received identical treatment without light exposure. Cell viability was assessed at 24–72 hours postirradiation using the CCK-8 assay or the CellTiter-Glo luminescent assay. IC₅₀ values were calculated by nonlinear regression using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA). All experiments were performed in triplicate and repeated at least 3 times independently.

High-Content Screening and Confocal Microscopy

Intracellular trafficking was evaluated using high-content screening (HCS) and confocal microscopy. For HCS, A431 cells were incubated with fluorescently labeled cetuximab-I21 at 4 °C (surface binding) or 37 °C (internalization) for 0–8 hours. Images were acquired using an Operetta CLS high-content imaging system (PerkinElmer, Waltham, MA). For confocal colocalization, cells were costained with LysoTracker Green DND-26 (75 nM) for lysosomal visualization and Hoechst 33342 (2 μg/mL) for nuclear counterstaining. Live cells were imaged using a Zeiss LSM 880 confocal microscope. Colocalization analysis was performed using ImageJ with the Coloc2 plugin, calculating Pearson correlation coefficients and Manders overlap coefficients.

Mitochondrial Membrane Potential Assessment

Mitochondrial membrane potential (Δψm) was measured using JC-1 dye, which forms red fluorescent J-aggregates in polarized mitochondria and green fluorescent monomers in depolarized mitochondria. A431 cells were treated with cetuximab-I21-3 (5 μg/mL), allowed to internalize for 4 hours, and irradiated with 50 J/cm². At 0, 1, 2, 3, and 4 hours postirradiation, cells were incubated with JC-1 (10 μg/mL) for 20 minutes at 37 °C. Fluorescence was measured using flow cytometry (BD FACSCelesta) with the FL1 channel (green monomers) and the FL2 channel (red J-aggregates). The red:green fluorescence ratio was calculated as an indicator of mitochondrial membrane potential.

Apoptosis Detection

Apoptosis was quantified using Annexin V-FITC/PI double staining. A431 cells were treated with cetuximab-I21-3 at various concentrations (0, 1.25, 2.5, 5, 10 μg/mL), irradiated with 50 J/cm², and cultured for 24 hours. Cells were harvested, washed with PBS, resuspended in binding buffer, and stained with Annexin V-FITC (5 μL) and PI (5 μL) for 15 minutes at room temperature in the dark. Samples were analyzed by flow cytometry within 1 hour. Early apoptotic cells were defined as Annexin V+/PI−, whereas late apoptotic/necrotic cells were Annexin V+/PI+.

In Vivo Biodistribution

Biodistribution studies were performed in A431 tumor-bearing female B-NDG mice (Biocytogen, Beijing, China). When tumors reached ∼100 mm³, mice received cetuximab-I21-2 (15 or 30 mg/kg) via tail vein injection. Live fluorescence imaging was performed using an IVIS Spectrum system (PerkinElmer) with excitation at 648 nm and emission at 651 nm at 0, 1, 4, 8, 24, 48, 72, and 96 hours postadministration. Regions of interest were drawn around the tumor and quantified using Living Image software. At 96 hours, tumors and organs were harvested for ex vivo imaging to confirm tissue distribution.

In Vivo Antitumor Efficacy

Antitumor efficacy was evaluated in A431 xenograft models using 2 study designs. In study 1, female B-NDG mice bearing A431 tumors (~85 mm³) were randomized into groups (n=6) receiving a saline control or cetuximab-I21-2 at 35.31 mg/kg, followed by single irradiation at 100 or 200 J/cm². In study 2, mice received saline, free I21, cetuximab-I21-2, or cetuximab-I21-3 at the indicated doses, followed by the irradiation schedules shown in Table 3. For cetuximab-I21-3, the most active regimen used 200 J/cm² irradiation at 6, 30, and 54 hours postdose. Light was delivered using a 660 nm LED source. Tumor volumes were measured 3 times weekly using the formula: volume (mm³)=(length×width²)/2. Tumor growth inhibition (TGI) was calculated as TGI (%)=[1 − (T/C)]×100.

TABLE 3.

Antitumor Efficacy and Tolerability of Cetuximab-I21 Constructs With Triple Light Irradiation in A431 Xenograft Model (Study 2)

Group	Treatment	Volume (mm3)	TGI-V (%)	Weight (mg)	TGI-W (%)	Mortality
G-1	Saline control	1902±165	—	2175±494	—	0/6
G-2	Saline + light (50 J/cm2×1)	1652±156	13.10	1890±445	13.10	0/6
G-3	Free I21 + light (50 J/cm2×1)	1248±96	34.35	1428±287	34.35	1/6
G-11	Cetuximab-I21-2 + light (200 J/cm2×3)	1480±142	22.18	1692±399	22.18	0/6
G-13	Cetuximab-I21-3 + light (200 J/cm2×3)	943±221***	50.42	1065±328***	51.09	2/6

Data presented as mean±SEM (n=6 per group).

Comparative antitumor efficacy of cetuximab-I21-2 and cetuximab-I21-3 was evaluated in A431 xenograft-bearing mice following a single intravenous dose and 3 light irradiations (200 J/cm²×3 at 6, 30, and 54 h postdose; 660 nm). Tumor volume and tumor weight were measured at the day 23 endpoint. Tumor growth inhibition based on volume (TGI-V) and weight (TGI-W) is shown. Mortality during the study period is indicated.

^*** P<0.001 versus saline control (G-1) by one-way ANOVA with Tukey post hoc test.

TGI indicate tumor growth inhibition.

Statistical Analysis

Data are presented as mean±SE of the mean (SEM) for in vivo studies and mean±SD for in vitro studies. Statistical comparisons between 2 groups were performed using an unpaired Student t test. Multiple group comparisons were analyzed by one-way ANOVA followed by the Tukey post hoc test. A P<0.05 was considered statistically significant. All analyses were performed using GraphPad Prism 9.0.

Ethics Statement

All animal experiments were conducted in accordance with institutional guidelines and approved by the Institutional Animal Care and Use Committee (IACUC) of InnoStar Bio-tech Nantong Co, Ltd, Jiangsu Province, People’s Republic of China. Two separate protocols were approved for this research: Protocol NN25337FP01 (IACUC Approval Number: IACUC-2025-m-781, approved August 27, 2025) and Protocol NN25337FP02 (IACUC Approval Number: IACUC-2025-m-876, approved September 19, 2025). The Animal Use License Number for the InnoStar Bio-tech Nantong facility is SYXK(Su)2025-0036, issued by Jiangsu Provincial Science and Technology Commission. The animal facility maintains AAALAC accreditation. All procedures were performed in strict accordance with the Guide for the Care and Use of Laboratory Animals (8th edition, National Research Council, 2011), the Regulations on the Management of Laboratory Animals (State Science and Technology Commission of China, 2017 revision), and the Jiangsu Provincial Measures for the Management of Laboratory Animals (Jiangsu Provincial Laboratory Animal Management Association, 2008). Animals were housed at InnoStar Bio-tech Nantong Co, Ltd (Building A18, No. 100 Dongtinghu Road, Linjiang, Haimen, Jiangsu Province, 226133, P.R. China), an AAALAC International fully accredited facility (Accreditation Number: 001783, continuing accreditation verified March 23, 2023). Female B-NDG mice (6–8 wk old, NOD-Prkdcscid-IL2rgammanull strain) were purchased from Biocytogen (Beijing, China) and housed under specific pathogen-free (SPF) conditions with a 12-hour light/dark cycle and ad libitum access to sterilized food and water, with appropriate environmental enrichment provided.

RESULTS

Conjugate Characterization and Quality Attributes

Both cetuximab-I21-2 and cetuximab-I21-3 were successfully synthesized using cysteine conjugation chemistry. In response to the importance of structural disclosure for evaluating new photosensitizers in the photoimmunotherapy field, the structural and photophysical characterization of the I21 payload-linkers and cetuximab-I21 conjugates is summarized in Figure S2, Supplemental Digital Content 1, http://links.lww.com/JIT/B69. The I21-2 and I21-3 payload-linkers had molecular weights of 1242.4750 and 1270.5290 Da, respectively, and red-light absorption maxima of 650.25 and 655.15 nm in 1×PBS (pH: 7.4). After conjugation to cetuximab, the resulting ADCs had calculated molecular weights of ∼161.9 and 162.2 kDa, respectively, with conjugate absorption maxima of 652.70 and 655.15 nm. Both constructs used maleimide-mediated cysteine conjugation and achieved a DAR of 8. EGFR binding assays confirmed that both conjugates retained binding to A431 cells. Extinction coefficients and singlet oxygen quantum yields were not determined in the present study and are recognized as additional parameters requiring future characterization.

Potent Light-Dependent Cytotoxicity With Complete Absence of Dark Toxicity

Both conjugates demonstrated potent, strictly light-dependent cytotoxicity in EGFR-positive cells (Table 1). Critically, neither cetuximab-I21-2 nor cetuximab-I21-3 showed any detectable cytotoxicity in the absence of light activation across the entire concentration range tested (up to 50 μg/mL), with killing rates remaining below 12% even at the highest concentrations. This complete absence of dark toxicity confirms that the I21 payload requires light activation to exert its cytotoxic effect.

TABLE 1.

In Vitro Light-Dependent Cytotoxicity of Cetuximab-I21 Conjugates Across Tumor Cell Lines With Different EGFR Expression Levels

Cell line	EGFR status	Conjugate	Dark (no light)	1 J/cm2	5 J/cm2	50 J/cm2
A431	High	I21-2	NA	0.065±0.008*	0.051±0.006	—
A431	High	I21-3	NA	—	—	0.739–1.093
HCC-827	High†	I21-2	NA	0.263±0.031	0.141±0.019	—
Huh7	Low	I21-2	NA	NA	22.63±3.42	—
NCI-N87	Low	I21-2	NA	NA	>29,000	—
A549	Negative	I21-3	NA	—	—	NA‡

Data are presented as mean±SD from 3 independent experiments.

In vitro cytotoxicity (IC₅₀) of cetuximab-I21-2 and cetuximab-I21-3 was evaluated in tumor cell lines with high, low, or negligible EGFR expression under dark conditions or following light irradiation at indicated doses (1, 5, or 50 J/cm²; 660 nm). IC₅₀ values are reported in μg/mL and represent mean±SD from 3 independent experiments. NA indicates no measurable cytotoxicity within the tested concentration range. EGFR-mutant status is indicated where applicable.

^* IC₅₀ values in μg/mL.

^† EGFR-mutant cell line—not tested at this light dose.

^‡ Not achievable (no significant cytotoxicity).

Upon light activation, cetuximab-I21-2 showed measurable phototoxicity in A431 cells, with IC₅₀ values of 0.0649 μg/mL at 1 J/cm² and 0.0514 μg/mL at 5 J/cm². Cetuximab-I21-3 showed IC₅₀ values of 1.093 μg/mL (24 h) and 0.739 μg/mL (48 h) at 50 J/cm². The EGFR-dependence of cytotoxicity was confirmed by evaluating activity across cell lines with varying receptor expression levels. HCC-827 cells (EGFR-mutant) showed similar light-dependent sensitivity (IC₅₀ 0.141–0.263 μg/mL). In striking contrast, EGFR-low cell lines demonstrated markedly reduced sensitivity. Huh7 cells showed an IC₅₀ of 22.63 μg/mL—∼440-fold less sensitive than A431 cells. NCI-N87 cells showed IC₅₀ values exceeding 29,000 μg/mL. A549 cells (EGFR-negative) showed no significant cytotoxicity (P<0.0001 vs. A431 cells), demonstrating >400-fold selectivity.

Receptor-Mediated Internalization and Lysosomal Trafficking

High-content screening imaging provided detailed insights into the cellular uptake mechanism (Fig. 1A). When A431 cells were incubated with fluorescently labeled cetuximab-I21-2 at 4 °C, the conjugate exhibited characteristic peripheral membrane staining, confirming efficient binding to cell surface EGFR without internalization. Upon warming to 37 °C, the conjugate was progressively internalized over 8 hours, transitioning from membrane-associated fluorescence to punctate intracellular staining characteristic of endosomal/lysosomal compartments. Time-lapse high-content imaging further illustrates the dynamic process of cetuximab-I21 internalization from the plasma membrane to intracellular compartments (Video S1, Supplemental Digital Content 3, http://links.lww.com/JIT/B71). These imaging data support progressive internalization of cetuximab-I21 under permissive temperature conditions. However, quantitative percent internalization was not determined in the present study, and no matched 4 versus 37 °C phototoxicity assay was performed. This limitation is now acknowledged in the Discussion.

FIGURE 1.

Intracellular trafficking and localization of cetuximab-I21 conjugates in A431 cells. A, High-content screening images showing membrane binding at 4 °C and progressive internalization at 37 °C over 0–8 hours. Cetuximab-I21-2 (green); nuclei stained with Hoechst 33342 (blue). B, Confocal microscopy colocalization analysis with LysoTracker Red demonstrating peak lysosomal accumulation at 3 hours (Pearson r=0.81±0.04). Cetuximab-I21 (green); lysosomes (red); nuclei (blue); colocalization (yellow). C, Time course of Pearson correlation coefficients for lysosomal colocalization, showing a maximum at 3 hours followed by a decline indicating processing/redistribution. Data are presented as mean±SD (n=3). Scale bars: 20 μm.

Confocal microscopy with LysoTracker colocalization revealed time-dependent trafficking through the endolysosomal pathway (Figs. 1B, C). The Pearson correlation coefficient between cetuximab-I21 fluorescence and lysosomal marker increased from 0.45±0.08 at 1 hour to a maximum of 0.81±0.04 at 3 hours (Manders M1 coefficient=0.87). By 4 hours, colocalization decreased to 0.72±0.05, and by 6 hours to 0.58±0.07, potentially indicating payload processing, release, or redistribution. This trafficking pattern is consistent with the well-characterized receptor-mediated endocytosis pathway for anti-EGFR antibodies.

Photodynamic Activation Induces Rapid Mitochondrial Depolarization

JC-1 assays demonstrated progressive mitochondrial membrane depolarization following photodynamic treatment (Fig. 2A). Control cells and cells treated with cetuximab-I21-3 without light irradiation maintained high Δψm (red:green ratio of 4.2±0.3 and 4.0±0.4, respectively), confirming that the conjugate does not impair mitochondrial function without light activation. Following light activation, the red:green fluorescence ratio declined progressively: 3.1±0.2 at 1 hour, 1.8±0.3 at 2 hours, 0.9±0.2 at 3 hours, and 0.3±0.1 at 4 hours postirradiation (P<0.001 vs. control). This represents >90% loss of mitochondrial membrane potential by 4 hours, indicating collapse of the mitochondrial proton gradient.^26,27

FIGURE 2.

Mechanism of cetuximab-I21-mediated cell death in A431 cells. A, JC-1 mitochondrial membrane potential assay showing progressive depolarization following photodynamic treatment. Red fluorescence (JC-1 aggregates) indicates polarized mitochondria; green fluorescence (JC-1 monomers) indicates depolarized mitochondria. The red:green ratio declined from 4.2±0.3 (control) to 0.3±0.1 at 4 hours postirradiation, representing >90% loss of Δψm. B, Annexin V/PI flow cytometry analysis demonstrating dose-dependent apoptosis induction at 24 hours postirradiation. Early apoptotic cells (Annexin V+/PI−) increased from 2.1% (control) to 24.8% at 10 μg/mL. Treatment: cetuximab-I21-3 at indicated concentrations, 50 J/cm² light. Data are presented as mean±SD (n=3). ***P<0.001 versus control.

Dose-Dependent Apoptosis Induction

Annexin V/PI flow cytometry revealed dose-dependent apoptosis induction following photodynamic treatment (Fig. 2B). Control cells maintained high viability (>91% Annexin V−/PI−), with minimal early apoptotic (2.1%) or late apoptotic/necrotic (3.2%) populations. Upon light activation, apoptosis was induced dose-dependently. At 10 μg/mL cetuximab-I21-3 with 50 J/cm², early apoptotic cells reached 24.8% and late apoptotic/necrotic cells reached 18.7%, with viable cells declining from 92.5% to 52.4% (P<0.001). These data indicate apoptosis-associated cell death after light activation and are consistent with mitochondrial injury. At the 24-hour time point, however, a substantial fraction of cells remained viable, indicating that this assay should be interpreted as a mechanistic readout rather than a complete measure of therapeutic efficacy.^28,29

Tumor-Selective Biodistribution With Sustained Retention

In vivo fluorescence imaging demonstrated tumor-selective accumulation of cetuximab-I21-2 (Figs. 3A, B). Quantitative tumor fluorescence values over time at both 15 and 30 mg/kg dosing levels are summarized in Table S1, Supplemental Digital Content 2, http://links.lww.com/JIT/B70. Following administration at 30 mg/kg, tumor fluorescence was detectable as early as 1 hour postinjection, with peak accumulation at 1–4 hours (mean fluorescence: 4.31–5.70×10⁹ photons/s). Importantly, tumor fluorescence was sustained through 96 hours at ∼40%–45% of peak values. Background signal in nontumor tissues declined more rapidly, resulting in increasing tumor-to-background ratios over time. Ex vivo imaging at 96 hours confirmed preferential tumor accumulation compared with liver and other organs (Fig. 3C), with a tumor-to-liver ratio of 4.4±0.8.

FIGURE 3.

In vivo biodistribution of cetuximab-I21-2 in A431 tumor-bearing mice. A, Representative whole-body fluorescence images (Ex=648 nm, Em=651 nm) at 0–96 hours following single intravenous administration (30 mg/kg). White circles indicate tumor location. B, Quantification of tumor fluorescence intensity (total radiant efficiency) over time. Peak accumulation at 1–4 hours with sustained retention through 96 hours. C, Ex vivo imaging of the excised tumor and liver at 96 hours (day 4) demonstrates preferential tumor accumulation. Data are presented as mean±SD (n=3).

Light Dose-Dependent Antitumor Efficacy

Two independent efficacy studies established the relationship between light dose and therapeutic outcome (Fig. 4; Tables 2, 3). In study 1 (single irradiation, day 12 endpoint), vehicle-treated controls reached 793±164 mm³. Light alone (50 J/cm²) had no significant effect (TGI 1.18%, P=0.918). Cetuximab-I21-2 with 100 J/cm² produced only modest effects (TGI 9.46%, P=0.423). However, cetuximab-I21-2 with 200 J/cm² achieved significant tumor growth inhibition (395±127 mm³, TGI 50.12%, P=0.0009). Individual tumor volume measurements for representative study 1 groups are provided in Table S2, Supplemental Digital Content 2, http://links.lww.com/JIT/B70.

FIGURE 4.

Antitumor efficacy of cetuximab-I21 photoimmunotherapy in A431 xenograft models. A, Study 1: tumor growth curves for cetuximab-I21-2 with single irradiation at varying light doses (day 12 endpoint). A 200 J/cm² achieved 50.12% TGI (P=0.0009). B, Study 2: tumor growth curves for cetuximab-I21-2 and cetuximab-I21-3 with triple irradiation (200 J/cm²×3, Day 23 endpoint). Cetuximab-I21-3 achieved 50.42% TGI (P=0.0002). C, Body weight changes over the treatment period showing transient weight loss (6%–8%) with recovery by day 7. D, Fluorescence imaging showing photobleaching of tumor-localized cetuximab-I21 following light treatment, confirming photodynamic activation. Data are presented as mean±SEM (n=6). ***P<0.001 versus saline control.

TABLE 2.

Antitumor Efficacy of Cetuximab-I21-2 With Single Light Irradiation in A431 Xenograft Model (Study 1)

Group	Treatment	Tumor Volume (mm3)	TGI (%)	P	Sig.
G-1	Saline (no light)	793±164	—	—	—
G-2	Saline + 50 J/cm2	783±143	1.18	0.918	ns
G-10	Cetuximab-I21-2 (35 mg/kg) + 100 J/cm2	718±145	9.46	0.423	ns
G-11	Cetuximab-I21-2 (35 mg/kg) + 200 J/cm2	395±127	50.12	0.0009	***

Data presented as mean±SD (n=6 per group).

Antitumor efficacy of cetuximab-I21-2 was assessed in A431 xenograft-bearing mice following a single intravenous dose (35 mg/kg), and 1 light irradiation was performed 6 hours postdose at (660 nm) at the indicated energy levels. Tumor volumes were measured at the day 12 endpoint. Tumor growth inhibition (TGI) was calculated relative to saline-treated controls.

^*** P<0.001 versus saline control (G-1).

ns indicate not significant.

Statistical significance was determined by one-way ANOVA with Tukey post hoc test.

In study 2 (triple irradiation, day 23 endpoint), cetuximab-I21-3 with 200 J/cm²×3 achieved the best outcome: 943±221 versus 1902±165 mm³ control (TGI 50.42% by volume, P=0.0002; TGI 51.09% by weight). Treatment was generally well-tolerated with transient body weight loss (6%–8%) at days 2–5, recovering by days 7–9 (Fig. 4C). Individual body weight changes for each animal are detailed in Table S3, Supplemental Digital Content 2, http://links.lww.com/JIT/B70. Fluorescence imaging confirmed photobleaching of tumor-localized conjugate following light treatment (Fig. 4D), providing visual confirmation of photodynamic activation.

DISCUSSION

This study provides preclinical characterization of cetuximab-I21-2 and cetuximab-I21-3 as EGFR-targeted, light-activated antibody-photosensitizer conjugates. IR700-based NIR-PIT remains the clinical and mechanistic benchmark for this field. The present work was not designed as a direct head-to-head comparison with cetuximab-IR700 and should not be interpreted as demonstrating superiority over IR700-based NIR-PIT. Instead, our data define a mechanistically distinct cetuximab-I21 platform characterized by EGFR-dependent uptake, intracellular trafficking, mitochondrial depolarization, apoptosis-associated phototoxicity, tumor-selective fluorescence retention, and moderate light-dose-dependent antitumor activity.

The dual-selectivity mechanism of cetuximab-I21 combines EGFR-mediated receptor targeting with local light activation. This design provides spatial control of phototoxicity and explains the absence of detectable dark toxicity in vitro. A schematic overview of the proposed mechanism is provided in Figure S1, Supplemental Digital Content 1, http://links.lww.com/JIT/B69. Importantly, this mechanism appears distinct from canonical IR700-based NIR-PIT. IR700 conjugates have been reported to mediate rapid plasma membrane damage after light exposure, whereas the current data indicate that cetuximab-I21 undergoes receptor-mediated internalization and endolysosomal trafficking before light-induced mitochondrial injury.

The >400-fold selectivity ratio between EGFR-high and EGFR-low cells further demonstrates that receptor-mediated targeting is essential for activity. This selectivity ratio compares favorably to many conventional ADCs, which often show selectivity ratios of only 10–100-fold. The combination of receptor selectivity and light-dependent activation should provide dual protection against off-target toxicity, potentially enabling more aggressive dosing strategies than are possible with conventional ADCs.

The mechanistic studies provide important insights into the cellular pharmacology of cetuximab-I21 conjugates. The trafficking pattern observed—with peak lysosomal colocalization at 3 hours followed by a gradual decline—is consistent with the well-characterized endocytic pathway for anti-EGFR antibodies.^30,31 The pronounced mitochondrial membrane depolarization (>90% loss of Δψm) indicates that mitochondria are key targets of the ROS-mediated damage. Mitochondria are particularly susceptible to oxidative damage due to their high lipid content, critical electron transport chain components, and limited antioxidant capacity.^32,33

The observed cell-death phenotype should be interpreted cautiously. Cetuximab-I21 induced mitochondrial depolarization and apoptosis-associated cell death after light activation, as shown by JC-1 and Annexin V/PI analyses. This phenotype differs from the rapid membrane-disruptive mechanism reported for IR700-based NIR-PIT, which can cause acute loss of membrane integrity, release of damage-associated molecular patterns and tumor antigens, and immune activation.^23,34–36 Because apoptosis is generally less acutely inflammatory than membrane rupture or necrotic cell death, we do not conclude from the present data that cetuximab-I21 induces immunogenic cell death equivalent to IR700-based NIR-PIT. Immune activation was not directly evaluated in the present study because the in vivo experiments were performed in immunodeficient B-NDG mice. Future studies should directly compare cetuximab-I21 with cetuximab-IR700 using assays of membrane damage, DAMP release, calreticulin exposure, ATP release, HMGB1 release, antigen presentation, immune-cell activation, and checkpoint blockade synergy in immune-competent models.

The in vivo efficacy studies indicate that relatively high light fluence was required for robust cetuximab-I21 activity in the current A431 xenograft model. In study 1, 100 J/cm² produced only modest tumor inhibition, whereas 200 J/cm² achieved 50.12% tumor growth inhibition at the Day 12 endpoint. In study 2, the strongest response was observed with cetuximab-I21-3 combined with 200 J/cm²×3 irradiation, achieving 50.42% tumor growth inhibition by tumor volume and 51.09% inhibition by tumor weight. These findings support proof-of-concept antitumor activity but also identify light-dose requirement as a major limitation of the current formulation and treatment schedule. Compared with published IR700-based NIR-PIT studies, the requirement for high fluence and repeated irradiation suggests that further optimization of photosensitizer structure, DAR, linker design, drug-light interval, fluence rate, and light delivery geometry will be required before translational conclusions can be made.

The sustained fluorescence signal observed through 96 hours may provide flexibility for scheduling light delivery, but it also raises an important photosafety question. Systemic photosensitivity was not formally evaluated in this study. Future pharmacokinetic and safety studies should assess blood clearance, skin and ocular retention, nontumor tissue fluorescence, whole-body phototoxicity, and the duration of any required light-avoidance period after dosing.

Several limitations should be acknowledged. First, although the chemical structures and selected photophysical properties of I21-2 and I21-3 are now disclosed, a direct head-to-head comparison with cetuximab-IR700 was not performed. Therefore, relative potency, light-dose requirements, mechanism of cell death, immune activation, and translational advantages cannot be concluded from the present study. Second, quantitative percent internalization and matched 4 versus 37 °C phototoxicity experiments were not performed, limiting definitive conclusions about the degree to which internalization is required for cetuximab-I21 activity. Third, the in vivo studies were performed in immunodeficient mice, precluding assessment of antitumor immunity. Fourth, the antitumor effect was moderate; no survival endpoint was included, and relatively high light fluence was required. Fifth, sustained fluorescence retention raises the possibility of prolonged photosensitivity, which will require formal pharmacokinetic and phototoxicity evaluation. Finally, the extinction coefficient and singlet oxygen quantum yield were not determined and should be included in future photophysical characterization.

CONCLUSIONS

Cetuximab-I21-2 and cetuximab-I21-3 are EGFR-targeted, light-activated antibody-photosensitizer conjugates with disclosed I21 payload-linker structures and DAR of 8. The conjugates demonstrated light-dependent cytotoxicity in EGFR-positive cells, absence of detectable dark toxicity, receptor-dependent selectivity, intracellular trafficking with lysosomal accumulation, mitochondrial depolarization, apoptosis-associated cell death, tumor-selective fluorescence retention, and moderate light-dose-dependent antitumor activity in A431 xenografts. These findings support cetuximab-I21 as a mechanistically distinct photoactivated antibody conjugate rather than a demonstrated replacement for IR700-based NIR-PIT. Direct IR700 comparison, quantitative internalization analysis, immune-response assays, photosafety studies, survival endpoints, and light-dose optimization will be required to define its therapeutic and translational potential.

Supplementary Material

cji-49-268-s001.pdf ^{(60.8KB, pdf)}

cji-49-268-s002.pdf ^{(110KB, pdf)}

Download video file ^{(3.1MB, mp4)}