IRSN, Institut de radioprotection et de sûreté nucléaire

Search our site :


Contact us :

En Fr

Enhancing Nuclear Safety


Research programs

ROSIRIS program

Last update on March 2012


Launched by IRSN in July 2009, the ROSIRIS program aims to better understand the mechanisms giving rise to secondary effects in radiotherapy. Rosiris is a multidisciplinary program combining expertise in radiopathology, radiotherapy and physical dosimetry.





Context and objectives



Standard radiotherapy practices prescribe optimisation of the doses delivered to the targeted tumour volume so as to obtain the best therapeutic efficacy, while reducing “insofar as possible” the dose delivered to the healthy tissue surrounding the tumour and so ensuring the quality of the treatment. It should be recalled however that radiotherapy is accompanied by secondary effects due to the presence of normal tissues in the irradiation field. Some of these effects disappear spontaneously; others unavoidably appear. Radiation-induced toxicity to healthy tissues is not only a limiting factor in escalation of the dose that can be delivered to the tumour, but its severity can unfortunately affect the quality of life of cancer survivors, who are more and more numerous.






Arterioles of the submucous membrane in a test rat and after 15 weeks of irradiation (27 Gy). The arteriole wall has thickened and its lumen has highly reduced (neo-intimal hyperblasia, double arrow). © IRSN (HDR of Agnès François)



The technical aspect of radiotherapy has become vastly more complex in recent years with the appearance of new technologies derived from particle accelerators, capable of performing sophisticated irradiation based on complex ballistics. This technological development provides new benefits to patients, but reveals new risks.


There is still a manifest lack of knowledge on the onset and progression of secondary complications of conventional and innovative radiotherapies and on their associated mechanisms. This makes quantification of the associated risk difficult.




The objective of the ROSIRIS experimental research program is to improve understanding of the secondary effects of radiotherapy. Ultimately, this program should improve prediction of the appearance of complications in healthy tissues, allow for optimisation and potentially personalisation of radiotherapy protocols, and open up new approaches to prophylactic or curative treatment.




This program is developed in collaboration with the CNRS and INSERM.






Issues and approach




The ROSIRIS program consists of different phases which should progressively supplement knowledge in dosimetry and radiobiology on the scale of organs (the tissue level).




The existence of a causal link between early radiation-induced effects and later consequences in tissues is the working hypothesis of the first phase of the program. An attempt will be made to establish links between biophysical modelling of early radiation-induced events and their later consequences at the cellular and tissue level. Biophysical modelling necessitates going beyond the concept of dose, used up to the present to predict biological effects, and will need to emphasise characterisation of the interactions of radiation with the cell on the basis of “bio-descriptors” of the topology of energy deposits. The stochastic nature of the energy deposits on a scale smaller than that of the cell nucleus, as well as the biological future of these deposits, is studied. In practice, the objective of this study is to compare the measured spatial distribution of foci (fluorescent marking of the proteins involved in dealing with radiation-induced DNA double strand breaks) with the simulated spatial distribution of the energy deposits in tissue, cellular and subcellular structures for a given radiation quality.







Blood vessel of rectum submucous membrane, in healthy human tissues, and irradiated by a radiotherapy. Brown marking shows PAI-1 overexpression in the blood vessel intern wall.  © IRSN (thesis of Rym Abderrahmani)




Initial work is now underway on photonic irradiation as used today in the great majority of radiotherapy irradiations. Irradiation using heavy ion beams will be studied next in the context of the development of hadron therapy. To correctly evaluate the spatial distribution of the foci, a special experimental methodology has been defined, allowing analysis of several thousand nuclei and hundreds of thousands of foci for each condition studied.




A high-throughput microscopy system, a suitable computational infrastructure and high-performance software have been put in place to fulfil the need for bulk analysis of images and their statistical interpretation. The energy deposited in cell nuclei on the nanometre scale is calculated by Monte Carlo simulation using the Geant4 code. The initial results show that it is possible to identify a statistical link between biological measurements and physical simulations. In other words, it appears to be possible to associate the variability in density of foci in a population of nuclei with the variations in estimated density of energy deposits in volumes comparable to the cell nuclei (several hundreds of µm3). This could initially allow the degree of biological variability in signalling DNA damage induced by exposure to ionising radiation to be evaluated. In addition, evaluation of the biological consequences of this same ionising radiation in experimental models, from the most rudimentary (cells) to the most complex (tissues), necessitates going beyond the radiobiology concepts used up to the present, and will have to strive to take into account the biology of complex systems.




The biology of complex systems is the integrated study of the biological functioning of cells in a dynamic network that coordinates and maintains tissue homeostasis – equilibrium. The objective is to show that in this new approach it is possible to identify the molecular networks operating in the mechanisms of tissue complications observed after radiotherapy using, first of all, a simplified cellular model representative of the microvessels. Specifically, high-dose irradiation of the endothelial cells in microvascularisation leads to early molecular changes that give rise to cellular death or the establishment of a later pathological phenotype in these cells.






Tryptase marking of mastocysts - in brown - on human tissues out of radiation field (left images) and in a radiotherapy field for a rectum adenocarcinoma (right images). Top images in mucous-submucous membrane, down images in blood vessels of submucous membrane.


© IRSN (thesis of Karl Blirando)



The proteins involved and differentially expressed in the early response of the endothelial cell to irradiation are studied by broad-spectrum methods (proteomic analysis) to model the response and characterise the essential participants playing a role in initiation and progression of the pathological phenotype. This approach via the biology of complex systems will allow an initial model of the endothelial response to irradiation and the persistence of its dysfunction to be obtained. Subsequently, the correlation explored between early biological effects and later biological effects will be validated in vivo by conventional radiopathology approaches involving innovative preclinical transgenic animal models. Study of the in vivo contribution of the vascular compartment in initiation and progression of radiation-induced lesions is in fact confronted with a technological barrier. The possibility of switching off a gene in a specific cellular compartment with the Cre-lox technology opens new prospects, especially in the area of radiopathology. The physio-pathological functions and associated molecular pathways involved in the continuum between the acute response to irradiation and later toxicity will be identified by bringing into play these innovative transgenic animal models, notably genetically deficient (KO) “conditional tissue-specific” mouse models. In the past two years, it has been possible to construct an initial KO conditional mouse model specifically at the endothelial level for an initial participant identified by proteomic analysis (PAI-1). The hypothesis of an endothelium-dependant continuum between early tissue effects and later radiation-induced effects is now being tested, PAI-1 being used as a molecular agent potentially allowing this relation to be established. This basis for new biological and physical knowledge should allow an initial improvement in models predicting cellular response to irradiation to be proposed.




In the second phase, the ROSIRIS program should allow confirmation or invalidation of whether a causal link can be established between the topology of energy deposits and their long-term tissue consequences. If the causality hypothesis is verified it should be possible to develop an initial model predicting the complications of radiotherapy starting from the energy deposits. This model should be parametrisable to take into account the individual characteristics of patients using biomarkers. This model will be extended to different types of radiation qualities used in radiotherapy.




In the third phase, the ROSIRIS program should allow the predictive model to be consolidated and its robustness to be tested in other organs at-risk (lung, heart, CNS, etc.). Demonstration of the robustness of such a model should open new therapeutic prospects for prevention of radiotherapy complications by suitable prophylactic treatment.




Finally, the last phase of the ROSIRIS program will have the objective of testing the robustness of the model in humans in an operating configuration and in clinical practice. The biological markers of radiosensitivity and the biological markers predicting the severity of complications will thus need to be validated by clinical studies. Ultimately, provided with the patient’s own biological and anatomical parameters, this model should allow an individualised risk to be attributed to each patient operationally. These new tools will allow vectorised external radiotherapy protocols to be optimised and personalised while ensuring optimal therapeutic efficacy.



Send Print

Involved IRSN laboratories

Further information



Send to a friend

The information you provide in this page are single use only and will not be saved.
* Required fields

Recipient's email:*  

Sign with your name:* 

Type your email address:*   

Add a message :

Do you want to receive a copy of this email?