Carole D. Thomas

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

BACKGROUND Photodynamic therapy is an established cancer treatment in which a photosensitizing agent is activated by exposure to light thus generating cytotoxic reactive oxygen species that cause cellular damage. METHODS A new photosensitizer synthesized at Curie Institute was used to treat retinoblastoma xenografts in mice, a glycoconjugated meso substituted porphyrin derivative, that showed some retinoblastoma cell affinity. The longitudinal follow-up of the tumors was carried out by (23)Na MRI (without adding exogenous contrast agents) to map the extracellular compartment and to characterize cell packing. Two regimens were followed to target either blood vessels alone or blood vessels and cancer cells simultaneously. RESULTS AND CONCLUSIONS Only the protocol targeting both cancer cells and blood vessels effectively induces cellular death, confirmed by histology at the end of the experiment. Sodium MRI evidences a huge change in the cellular density of tumors only 24h after a double targeting (vascular and cellular) PDT treatment. We suggest that this change was possibly due to a bystander effect that can be promoted by the intercellular signaling favored by the high cellular density of retinoblastoma. These results indicate that non-invasive (23)Na imaging (which detects the tumor response to treatment from very early stages) in association with non-mutagenic therapies represents an effective option for tailored and individualized clinical treatments.

Morphological and carbogen-based functional MRI of a chemically induced liver tumor model in mice

A multifocal mouse liver tumor model chemically induced with 5,9‐dimethyl‐7H‐dibenzo[c,g]carbazole was investigated by respiratory‐triggered morphological and functional MRI (fMRI) at 4.7 Tesla. The model is characterized by the presence of two tumor types: hypovascular cholangioma and vascularized hepatocellular carcinoma (HCC). Growth curves measured by 3D‐MRI showed limited growth of cholangiomas and rapid growth of HCCs after a latency of about 25 weeks. Functional imaging based on T  2* ‐weighted fast gradient‐echo MRI and carbogen breathing was optimized for liver imaging in mice. A response to carbogen was observed in HCCs but not in cholangiomas. Transversal analysis (50 HCCs) of signal change upon carbogen revealed four different types of response patterns: 1) signal increase upon carbogen administration (74%); 2) small or insignificant signal change (10%), 3) transient signal decrease and delayed increase (8%), and 4) signal decrease (8%). Longitudinal follow‐up of a subgroup (N = 17) showed that an initially observed type 1 response, attesting to the presence of a functional vasculature, remained stable for at least 3 weeks in 14 HCCs. A switch from a type 1 response to another response type may be useful for demonstrating, in a noninvasive manner, a disturbance of tumor vasculature induced by anti‐vascular or anti‐angiogenic therapy. Magn Reson Med 50:522–530, 2003.

Relationship between tumour growth rate and carbogen-based functional MRI for a chemically induced HCC in mice

We previously performed MRI studies of HCC (hepatocellular carcinomas) in mice showing the feasibility of measuring a carbogen effect. In the present study carbogen response of the whole tumour was compared with growth characteristics during longitudinal follow-up. HCC were chemically induced. The imaging protocol at 4.7 T comprised a fast spin-echo sequence for high-resolution screening and measurement of growth curves, and a fast gradient echo sequence allowing an entire T2*w image acquisition per respiratory cycle to perform fMRI under carbogen breathing. A new parameter, T+, the fraction of tumour voxels with increased intensity under carbogen was measured on manually defined ROIs. Twenty-two HCC were followed for 3–10 weeks. Tumours were divided into two groups, “regularly” and “irregularly” growing tumours. A linear correlation between T+ and tumour growth rate was observed only for “regularly” growing HCC. These results suggest a link between tumour growth rates and tumour fractions exhibiting signal increase upon carbogen breathing. They are compatible with observations by others that rapidly growing tumours are more hypoxic than slowly growing ones. Combined measurement of T+ and tumour growth may become a useful noninvasive follow-up approach for assessment and/or management of therapies involving vasculature-targeting and anti-proliferative drugs.

PDT induced bystander effect on human xenografted colorectal tumors as evidenced by sodium MRI

BACKGROUND Previous in vivo studies on photodynamic therapy (PDT)-treated, high cellular density tumors showed evidences of a bystander effect accompanying the therapy, cellular death continuing beyond the limits of the photochemical reactions in time and space. This process is generated by the initially damaged cells on the light pathway. The aim of this study was to determine if the bystander effect may be induced as well in colorectal xenografted tumors (less compact structure) and if the cellular signaling depends primarily on cellular proximity or not. METHODS The photosensitizer was a glycoconjugated, meso substituted porphyrin derivative synthesized at Institut Curie. The longitudinal follow-up of the tumors was carried out by (23)Na/(1)H MRI, ideal imaging modality for mapping the extracellular compartment. Two regimens were followed in order to target either blood vessels alone or blood vessels and cancer cells simultaneously. RESULTS The antivascular PDT did not succeed to arrest the tumors growth at the end of the follow-up. For double targeting PDT, we managed to stop the tumoral evolution. Sodium MRI evidenced a bystander effect. CONCLUSION The results obtained showed that the bystander effect is more difficult to induce for the type of colorectal tumors used in this work. It needs a double treatment, 4 days apart, in order to be promoted.

Cellular density, a major factor involved in PDT cytotoxic responses: Study on three different lines of human retinoblastoma grafted on nude mice

BACKGROUND PDT represents a very localized and non-mutagen antitumoral treatment using a photosensitive molecule (porphyrin family) light activated. The first way of cell damage is a direct one, active on the very site where ROSs have been produced. The second one is indirect by activating and transmitting the processes of cellular death signaling. In order to seek for a better characterization of the photo-biology involved in in vivo PDT and to better understand the differences on the treatment outcome, we have used three different human retinoblastomas xenografted on mice. METHODS Mice were treated according to the double targeting protocol exposed in a previous paper. One i.v. dose (0.6 mg/kg) of PS was followed by a second dose, separated by a 3 h interval (double targeting PDT). As a consequence both cancer cells and blood vessels were targeted. The treatment was repeated two times, at 4 days interval. RESULTS First of all, sodium MRI revealed qualitative differences in the sodium average content of the three retinoblastoma lines before treatment. After the PDT treatments the tumor responses were different between the lines as revealed by sodium MRI and later on by histology. CONCLUSIONS We have put into evidence that PDT is accompanied by a bystander effect that may propagate the cellular death triggered by the initial photoreaction. This effect is highly dependent on the cellular density of the tissue; therefore this factor is to be taken into account in clinical PDT protocols.

Iterative Variance Stabilizing Transformation Denoising of Spectral Domain Optical Coherence Tomography Images Applied to Retinoblastoma

Background: Due to the presence of speckle Poisson noise, the interpretation of spectral domain-optical coherence tomography (SD-OCT) images frequently requires the use of data averaging to improve the signal-to-noise ratio. This implies long acquisition times and requires patient sedation in some cases. Iterative variance stabilizing transformation (VST) is a possible approach by which to remove speckle Poisson noise on single images. Methods: We used SD-OCT images of human and murine (LH Beta-Tag mouse model) retinas with and without retinoblastoma acquired with 2 different imaging devices (Bioptigen and Micron IV). These images were processed using a denoising workflow implemented in Matlab. Results: We demonstrated the presence of speckle Poisson noise, which can be removed by a VST-based approach. This approach is robust as it works in all used imaging devices and in both human and mouse retinas, independently of the tumor status. The implemented algorithm is freely available from the authors on demand. Conclusions: On a single denoised image, the proposed method provides results similar to those expected from the SD-OCT averaging. Because of the friendly user interface, it can be easily used by clinicians and researchers in ophthalmology.