In vivo efficacy of photodynamic therapy in three new xenograft models of human retinoblastoma

https://doi.org/10.1016/j.pdpdt.2010.09.003Get rights and content

Summary

Purpose

Retinoblastoma is the most common primary intraocular tumor in children. In industrialized countries, 95% of patients are cured by chemotherapy and conservative treatments. However, these treatments can increase the risk of secondary tumors in patients with a constitutional alteration of the RB1 gene.

Photodynamic therapy represents a nonmutagenic therapeutic approach, and may reduce the incidence of secondary tumors.

To study the in vivo efficacy of photodynamic therapy, human retinoblastoma xenografts were established in nude mice.

Methods

Three xenografted cell lines, RB102-FER, RB109-LAK and RB111-MIL, were characterized and used for therapeutic evaluation.

Mice were randomly divided into control and treatment groups with 5–8 mice in each group. Treatment groups received irradiation alone, photosensitizer alone or both in 2 of the 3 models and in the third model, photosensitizer plus irradiation was compared to untreated controls. mTHPC was injected intraperitoneally at a dose of 0.6 mg/kg and verteporfin intravenously at a dose of 1 mg/kg. Illuminations were performed 24 h after mTHPC and 1 h after verteporfin injections.

Results

A transient but significant response to mTHPC was observed for RB102-FER (p = 0.03) and a significant response to mTHPC for RB111-MIL (p < 10−4) with partial regression maintained for more than 60 days. No significant difference between the different groups was observed for RB109-LAK, except in the verteporfin plus laser group (p = 0.01).

Conclusions

The studies confirmed the suitability of the three xenograft models for the evaluation of photodynamic therapy in retinoblastoma. Our findings suggest that PDT may represent an alternative conservative treatment for these tumors.

Introduction

Retinoblastoma (RB) is a rare disease, but is the most common primary intraocular tumor of childhood [1] with an incidence of 1 in 15,000–20,000 births. In about 40% of patients, the disease is bilateral and is associated with a mutation of the retinoblastoma gene, RB1, localized on chromosome 13, band q1.4 [2]. This mutation is also present in 15% of patients with unilateral retinoblastoma.

While the vital prognosis related to RB itself is excellent, with cure rates greater than 95% in industrialized countries, long-term survival in patients with the RB1 gene mutation is reduced, due to the risk of secondary malignant neoplasms occurring years after treatment, which represent the leading cause of death in patients with hereditary RB [3], [4].

The current therapeutic options available to treat patients with RB include radical treatment by enucleation or conservative treatment using local thermotherapy [5], [6], cryotherapy [7] or brachytherapy [8]. Primary chemotherapy is often necessary to reduce tumor size and make the tumor accessible to conservative treatment [5], [9], [10], [11], [12]. The use of external-beam radiotherapy is increasingly restricted due to the risk of late effects, including secondary sarcoma [13]. Chemotherapy with carboplatin also has potential risks of late effects, including increased risk of second cancer in patients with RB1 gene mutation, as well as short-term side effects. In this context, nonmutagenic therapies such as photodynamic therapy have considerable potential, particularly in patients with hereditary RB.

Photodynamic therapy (PDT) is already used as a therapeutic modality in the treatment of certain types of cancer and involves the use of a light-sensitive chemical agent, or photosensitizer, that when exposed to light, generates cytotoxic reactive oxygen species, leading to cell death through apoptosis or necrosis [14], [15].

The mechanisms of PDT-induced cell damage involve a complex interplay between the tissue, the photosensitizer and the light [16]. Adequate amounts of photosensitizer and light must be delivered to the target tissue with a sufficient supply of oxygen. PDT-induced cell damage also depends on the subcellular site of the photosensitizer [17]. PDT targets tumor cells, as well as the microenvironment, including normal and tumor vascularization and immune cells [18]. The efficacy of PDT in RB was demonstrated in early experimental studies using intraocular xenograft RB, however ocular side effects, such as intraocular hemorrhage and reversible cloudy cornea, were observed [19], and clinical studies were discontinued [20]. A possible explanation for these side effects may be that the photosensitizer used, Photofrin®, was less specific and less effective than those used today, requiring a high intratumoral concentration and a high dose of light exposure. More recent, less toxic photosensitizers such as m-tetrahydroxyphenylchlorin (mTHPC) and verteporfin have opened up new possibilities, however more extensive testing in preclinical studies is needed before these agents can be used to benefit patients with RB.

The in vitro models used to study the efficacy of PDT in RB have traditionally relied on two main cell lines, Y79 [21] and WERI Rb1 [22], and the newer photosensitizers have been shown to induce in vitro cell death in the Y79 RB cell line [23], [24]. Although these cell lines are useful for in vitro screening of new therapeutic compounds, Y79 presents a MYCN amplification that is observed in only a few retinoblastomas [25], [26], and this cell line may therefore not be representative of RB. Moreover, contamination problems have been described with Y79 raising doubts about the genetic stability of this cell line [27].

A recent study has described the use of three additional cell lines, RB247C3, RB355 and RB383, to investigate the in vitro effects of verteporfin, a second-generation photosensitizer, in PDT [28].

The animal models currently available to study RB are of two types: xenografts [29], [30], [31], [32] or genetically modified mouse strains [33]. One such xenograft model involves intraocular injection of the Y79 cell line in newborn rats [34]. The tumor cells invade the vitreum within two weeks and present most of the clinical and histological characteristics of human RB, however, because it relies on injection of a cultured, homogeneous cell population, this model remains artificial.

In recent years, much interest has focused on the use of transgenic mouse models, and several new preclinical models of RB have been developed [31]. Recently, a non-chimeric knock-out mouse model of RB was generated. This model entails the introduction of 6 different alleles and inactivation of RB1, p107 and p53 genes [33]. This model is of particular interest in studying tumorigenesis in RB and represents a major advancement compared to the first genetically engineered mouse strains with an inactivated RB1 that did not develop RB [35], [36], [37]. This model may not however be well suited to the study of the effects of PDT in RB since p53 is known to play a role in the apoptosis induced by photoactivation of mTHPC [38].

In an attempt to develop a preclinical model particularly well suited to the study of PDT under conditions that closely recapitulate the clinical environment of human RB, we have generated three new xenografted mouse models using tissue obtained directly from surgical samples. The validity of the xenograft tumors as a model for RB was tested by studying the response to one of the standard chemotherapy regimens used in RB, i.e. carboplatin and etoposide. The models were then used to investigate the efficacy of PDT with two of the newer, second-generation photosensitizers, mTHPC and verteporfin, in the treatment of RB.

Section snippets

Clinical characteristics

With the parents consent, surgical samples were taken from the enucleated eyes of 14 patients, 3 boys and 11 girls, with a median age of 2.4 years (range: 4 months–5 years). Three of the fourteen tumors gave rise to xenograft tumors: RB102-FER, RB109-LAK, and RB111-MIL. All three patients presented unilateral, non-familial RB.

Constitutional RB1 status of the patients

Constitutional analysis of the RB1 status of the three patients from whom the xenograft tumors had been obtained was performed. A constitutional mutation was identified in

Histologic analysis of xenografted tumors

Three out of fourteen tumors were established in nude mice (21.4%).

Each primary tumor was analyzed and classified according to current guidelines. Two of the three tumors that grew in mice were classified as poorly differentiated RB (Fig. 1a), and one was classified as moderately differentiated RB (Fig. 1c). After grafting into nude mice, the tumors presented an undifferentiated pattern (Fig. 1b and d).

Discussion

We performed a xenograft protocol using 14 different retinoblastomas in immunocompromized mice. Three new models of RB were established from these tumors xenografted into nude mice. We confirmed that the xenografts retained the genomic characteristics of RB [26], [44], [45]. This stability between the primary tumor and the xenograft, as well as their sensitivity to the carboplatin–etoposide combination, suggest that our models present many of the characteristics of clinical RB and may therefore

Acknowledgments

Dr. I. Aerts was supported for this work by La Fondation pour la Recherche Médicale (Grant FRM: DEA20040901706). L’Association pour la Recherche sur le Cancer financed the laser diode Spectra physics (Grant no. 7810). The authors thank la Fondation de l’Avenir, Retinostop and l’Institut Curie for their financial support and Quantel Medical for the loan of laser.

References (52)

  • I. Laville et al.

    Synthesis, cellular internalization and photodynamic activity of glucoconjugated derivatives of tri and tetra(meta-hydroxyphenyl)chlorins

    Bioorg Med Chem

    (2003)
  • I. Laville et al.

    A study of the stability of tri(glucosyloxyphenyl)chlorin, a sensitizer for photodynamic therapy, in human colon tumoural cells: a liquid chromatography and MALDI-TOF mass spectrometry analysis

    Bioorg Med Chem

    (2004)
  • O. Bourdon et al.

    A comparative study of the cellular uptake, localization and phototoxicity of meta-tetra(hydroxyphenyl) chlorin encapsulated in surface-modified submicronic oil/water carriers in HT29 tumor cells

    J Photochem Photobiol B

    (2000)
  • M. Lupu et al.

    23Na MRI longitudinal follow-up of PDT in a xenograft model of human retinoblastoma

    Photodiagnosis Photodyn Ther

    (2009)
  • D.H. Abramson et al.

    Update on retinoblastoma

    Retina

    (2004)
  • S.H. Friend et al.

    A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma

    Nature

    (1986)
  • F.L. Wong et al.

    Cancer incidence after retinoblastoma. Radiation dose and sarcoma risk

    JAMA

    (1997)
  • A.L. Murphree et al.

    Chemotherapy plus local treatment in the management of intraocular retinoblastoma

    Arch Ophthalmol

    (1996)
  • J.A. Shields et al.

    Cryotherapy for retinoblastoma

    Int Ophthalmol Clin

    (1993)
  • G.L. Chantada et al.

    Experience with chemoreduction and focal therapy for intraocular retinoblastoma in a developing country

    Pediatr Blood Cancer

    (2005)
  • C. Levy et al.

    Role of chemotherapy alone or in combination with hyperthermia in the primary treatment of intraocular retinoblastoma: preliminary results

    Br J Ophthalmol

    (1998)
  • C. Rodriguez-Galindo et al.

    Treatment of intraocular retinoblastoma with vincristine and carboplatin

    J Clin Oncol

    (2003)
  • C.L. Shields et al.

    Chemoreduction for retinoblastoma may prevent trilateral retinoblastoma

    J Clin Oncol

    (2000)
  • R.A. Kleinerman et al.

    Risk of new cancers after radiotherapy in long-term survivors of retinoblastoma: an extended follow-up

    J Clin Oncol

    (2005)
  • T.J. Dougherty et al.

    Photodynamic therapy

    J Natl Cancer Inst

    (1998)
  • J. Winther et al.

    Experimental studies of photodynamic therapy in a retinoblastoma-like tumour

    Acta Ophthalmol Suppl

    (1987)
  • Cited by (0)

    1

    Contributed equally to the work.

    View full text