Predictive test for radiotherapy induced reactions: women at high risk

Dr. Annette LEXA (PhD Toxicology)

November 20, 2017

While mainstream media give the impression of discovering a hot scoop and a possible polemic [1] , the French[2] and international biomedical community is now perfectly aware of the limits of breast cancer screening and its burden (limited benefits, imperfect sensitivity and specificity [3], over-diagnosis, “preventive” surgery, morbidity and mortality by radiotherapy) but does not communicate this fact to the general public.

Cancer Rose with FORMINDEP (french non profit organization for independent medical education and information), with several independent general practitioners on social networks (Jean Claude Grange, Claudina Michal-Teitelbaum, JB Blanc, Dominique Dupagne, Jaddo, etc.) and several women (Martine Bronner, Rachel Campergue, etc.) have simply relayed – voluntarily, independently and free of charge – the scientific information available to the first concerned, work which should normally have been done by professionals and which has not been done for 2 main reasons : some have severe cognitive biases, others are very afraid of the “feelings” of women who might lose “confidence” in screening, which must therefore be improved as soon as possible. This probably explains this first Pink October 2017, which is strangely silent.

Regarding the risk associated with radiotherapy, Cancer Rose had already alerted women to the risk [4]. However, it continues to be largely underestimated [2]. Following an e-mail sent to us by an anxious woman on October 4, we looked for an update on the predictive tissue radiosensitivity tests available in France [5]. And what we discovered is not very pleasant, we should say again.

Radiosensitivity is a real public health problem

60% of the 380,000 new cases of cancer in France are treated by radiotherapy. Patients receive sessions of the order of 2 Gy [6], repeated daily. Even with state-of-the-art equipment, the risk of irradiating healthy connective tissue [7] is unavoidable.

 Among patients treated by radiotherapy, 5 to 20% may show adverse reactions such as dermatitis, fibrosis, rectitis (this is called tissue radiosensitivity), secondary radiation-induced cancers [8] (between 5 and 12%) (this is called cancer radio-susceptibility).  At present, no personalized medicine currently takes into account the individual risk of radiosensitivity or radiosusceptibility. Moreover, international radiation protection rules (e.g. Sievert[6] still consider individuals as being all radioresistant for these two concepts. Even nowadays, people are exposed to the same dose of radiation that can lead either to a cure without side effects, a cure with more or less serious to fatal side effects in the case of coronary artery disease, or even death in 100% of exceptional cases of well-characterized rare genetic diseases. All treatment parameters are set without taking into account individual variations. Therefore, a person who will have a 10 times higher risk of developing cancer, will still receive the same dose as a person repairing his DNA lesions well and is unlikely to develop secondary cancer.

Indeed, the biggest problem is the double-strand breakage of the DNA (which we will call DSB, it is the most serious lesion that can be suffered by the DNA because both copies are affected). Normally the repair of DSB is initiated by a cytoplasmic protein called ATM. For about fifteen years, the Radiobiology Group of UMR 1052 INSERM (Lyon) has been working to understand this mechanism and to develop an individual predictive test.

This protein is activated for any oxidative stress producing DSBs and passes into the nucleus to trigger DNA repair. This group of researchers has identified 3 groups of humans:

– Group I = radioresistant, low-risk cancer patients who quickly and correctly repair DSBs, which represent about 75-85% of the population.

– Group II = moderate radiosensitive, and radiosusceptible (high risk of cancer), who repair late and with errors even at low radiation doses. Individuals in Group II would represent 20% of the general population. When these individuals are subjected to radiotherapy sessions, they can more easily trigger radiation-induced cancers, but also dermatitis, rectitis, coronaritis, etc.

– Group III = hyper-radiosensitive individuals who do not repair the DSB and are highly radiosusceptible, these present rare genetic syndromes diagnosed early in life (ataxia telangiectasia) and represent 1 person/100,000 but 0.5 to 5% of individuals worldwide would be carriers of a mutation of the ATM gene.

This classification also exists for tumor cells, independently of the group of healthy tissues (e.g., a group II patient may show group I, II or III tumors).

Finally, there is the problem of the repetition of low doses that can overwhelm the restorative capacity of the irradiated cells. This risk must be taken into account for young patients at high family risk. It is also known today that people with the BRCA1 or BRCA2 mutation have a risk associated with a certain radiosusceptibility, as well as people carrying mutations on p53 or ATM.

In addition to the risk associated with radiotherapy, the risk associated with radiodiagnosis should not be overlooked.  It accounts for the majority of medical exposures, usually to healthy individuals. If the exposure dose is 100 to 1000 times lower than in radiotherapy, the average exposure is constantly increasing : it is justified to question the safety of low doses of irradiation in group II patients, and in particular in young patients carrying the mutation ATM (1% incidence), or BRCA1 (1/1000 incidence) and BRCA2 (1/2000), the latter mutations presenting for the carrier an excess risk of breast cancer by a factor of 5 to 10 (this represents nearly 1% of the general population which has, for women, a risk of breast cancer 5 to 10 times higher than normal). Mammography generates 2 x 2 mGy images, spaced 3 minutes apart. Repeated doses induce CDBs and may induce additional DSBs during repair, thus increasing the effect of low doses. Studies have shown a risk of radiation-induced breast cancer for cumulative doses of 100 mGy. This is one of the reasons why mammograms should not be cumulated too much before the age of 50 and especially not before the age of 30 for women in group II.

Radiosensitivity tests: ATM transit test and apoptosis test 

In order to be truly predictive, a test must meet 3 criteria:

– It must be founded on a solid scientific basis

– it must have been tested on a large number of individuals of different radiosensitivity

– it must show the highest possible statistical power, independent of the dose, the nature of the tissue response and the location of the tumor.

Experience has shown that the more rapid the test, the less powerful it is.

Generally based on non-irradiated cells, proteomic and genomic approaches are currently not convincing and currently no genetic marker or gene expression can claim to predict radiosensitivity.

The ATM transit test

However, a simple test exists today to analyze the activity of the ATM protein in the nucleus during a 2 Gy irradiation of skin fibroblasts taken from the person. This test was developed by the radiobiology group of UMR1052 INSERM (French National Institute of Health and Medical Research) in Lyon. The test is based on the principle that the slower the transit of the ATM protein in the nucleus, the more radiosensitive the patient is. This test can distinguish 5 clinical grades of severity of tissue reactions (equivalent to degrees of burn), its sensitivity is 100% and its specificity 92%.

The development of the ATM test was carried out as part of a larger research project called the COPERNIC project [9]. Two tests will be marketed at the beginning of 2018 by NEOLYS DIAGNOSTICS[10] in several forms: a rapid test[11] (performed in an anti-cancer center, results in 4 to 5 hours) called “first sorting” from a rapid skin or blood sample for a few hundred euros, and a complete test* from a cutaneous sampling (3 weeks). The radiobiology group of the UMR1052 Inserm and the company NEOLYS DIAGNOSTICS collaborate, within a framework set by the Inserm deontology commission, to better define the regulations and ethics regarding radiosensitivity issues in order to give a clear legal framework to the use of these tests, which are always prescribed by doctors.

ATM transit theory also finds applications for toxicity related to non-radiative agents such as metals or pesticides: tests are being developed in this direction, which requires absolute rigor in the management of information and the implementation of alternatives when they would prove to be positive.

*In fact the price of the characterization test has not yet been finalized. It should be around 2,000€ (instead of 1,500€ as indicated) for the 3 weeks required for this test in a specialized medical analysis laboratory (excluding the anti-cancer center). Thanks to Mr. Gilles Devillers – pharmacist – president of Neolys Diagnostics, for the clarifications made.

The apoptosis test

There is another test model based on the apoptosis[12] of lymphocytes circulating in the blood, irradiated at 8 Gy. But this model is only valid for lymphocytes (which have been shown not to be the best candidates to represent breast connective tissue) and, moreover, the dose is higher (remember that a radiotherapy session exposes to a dose of 2 Gy). Originally, this test is only valid for delayed tissue reactions (such as those observed in some prostate cancers). It is based on the premise that more apoptosis means more radioresistance, which is contrary to the consensus observation that the more radio-induced death is observed, the more radiosensitive the cells are.

In addition, this test is based on the premise that lymphocyte death can predict the death of other types of tissue, such as conjunctive tissue, which rarely apoptoses.

The apoptosis criterion is all the more problematic as the test uses a high dose of 8 Gy (there is systematically a rapid apoptosis in 24 to 48 hours at this high dose).

At 2 Gy, there is less apoptosis and moreover it is observed at this dose that the more extensive the apoptosis, the more radiosensitive the patients are (Baijer et al, 2016 ). The figures put forward suggest that the test based on lymphocyte apoptosis shows lower statistical performance than the ATM test. Finally, the recruitment of patients for the development of the test can be problematic: the prospective aspect [13] of the trials favors rather radioresistant subjects and the performance of the test to detect radiosensitive patients is therefore based on a few dozen patients, which can introduce an undiscussed bias.

 This test is currently being developed by the company NOVAGRAY [14]. On its website, the company proposes to adapt the dose and the sequence of the sessions “to the woman patient’s profile” when the patient has agreed to pay 1500 euros for this test. Evoking “a woman patient” leaves one wondering: since the test was initially developed for prostate cancer, the breast not being explicitly mentioned, is the woman more willing to buy a test than a man patient? This is all the more surprising since the site states that tests for prostate and lung cancer are in development. This suggests that the test currently available at NOVAGRAY is not specific for breast cancer.


There is little clear and accessible information on this type of personalized medicine for women facing breast cancer screening and treatment. And yet, it is vital for them. Patients already weakened by the announcement of the disease and the care pathway to be undertaken cannot alone assume the choice of a radiosensitivity test. Does the doctor prescribing this non-reimbursed test provide them with balanced and up-to-date information?

How can they discern between apoptosis and ATM testing for a comparable price? Should cheaper first-sort tests be favored? Moreover, while radiosusceptibility tests could be developed through the ATM transit theory, apoptosis (and therefore cell death) tests do not meet patients’ expectations regarding mammographic exposure.

We are facing an extremely serious inequality of access to information and innovation, while lives are at risk. Despite the very complex mechanisms of individual response to radiation, there is now an urgent need for the authorities to take into account and better frame all issues related to individual radiosensitivity and radiosusceptibility.


– Influence of Nucleoshuttling of the ATM Protein in the Healthy Tissues Response to Radiation Therapy: Toward a Molecular Classification of Human Radiosensitivity , The COPERNIC project investigators,
 Int J Radiation Oncol Biol Phys, Vol. 94, No. 3, pp. 450- 460, 2016

– The Henri Mondor Procedure of Morbidity and Mortality Review Meetings: Prospective Registration of Clinical, Dosimetric, and Individual Radiosensitivity Data of Patients With Severe Radiation Toxicity , Yazid Belkacemi et col. , Int J Radiation Oncol Biol Phys, Vol. 96, No. 3, pp. 629-636, 2016

– TNFSF10/TRAIL regulates human T4 effector memory lymphocyte radiosensitivity and predicts radiation-induced acute and subacute dermatitis, Baijer et col., Oncotarget. 2016 Apr 19;7(16):21416-27

– Radiation-induced CD8 T-lymphocyte Apoptosis as a Predictor of Breast Fibrosis After Radiotherapy: Results of the Prospective Multicenter French Trial, David Azria et col. , EBioMedicine,  December 2015 Volume 2, Issue 12, Pages 1965–1973

 – Low radiation doses: towards a new reading of risk assessment? Anne-Fleur Perez, Clément Devic, Catherine Colin, Nicolas Foray , Bull Cancer 2015

– Individual radio sensitivity: an old notion and its future , Conclusions of the December 16, 2013 seminar organized by the ASN , 2014

– Radiosensitivity Evidence of an individual factor, Nicolas Foray, Catherine Colin and Michel Bourguignon, Medicine/Science 2013; 29: 397-403

– 100 years of individual Radiosensitivity: How We Have Forgotten the Evidence, Nicolas Foray, PhDCatherine Colin, MD, PhD Michel Bourguignon, Radiology: Volume 264: Number 3—September 2012

– DNA double-strand breaks induced by mammographic screening procedures in human mammary epithelial cells , Catherine Colin et col. , Int J Radiat Biol. 2011 Nov;87(11):1103-12.

 [1]Breast cancer: the relevance of screening in question, Catherine Ducruet, Les, 07/10/17 –

[2] Delaloge S, et al. Breast cancer screening: on the way to the future. Bull Cancer (2016), 10.1016/j.bulcan.2016.06.005

[3] The sensitivity of a test is its ability to detect as many ” ill people ” as possible (false negatives are avoided). Specificity is its capacity to detect only ill people (avoid false positives).

 [4] Mammograms and radiosensitivity, Annette LEXA ,

[5]  “Being very anxious about the effects of radiotherapy, I discovered a company (Novagray) that markets a radiosensitivity test. In April 2017 this test was not reimbursed and costs 1500 euros. I finally did not accept this test for financial reasons but also because in case of non-radiosensitivity the number of sessions is decreased and the dose increased, which did not seem reassuring to me. Do you know this test and can you give me your opinion? My radiotherapy sessions are over but I refuse the control mammogram, considering that there is no consideration of all the accumulated radiation. »

[6]  The sievert (Sv) is an international unit used to give an assessment of the biological impact of radiation on humans. The Gray (Gy) is the absolute dose of radiation received/Sievert. A dose of 2 Gy is about 1000 times the dose received for a mammogram image.

[7] Connective tissue is the supporting tissue (collagen fiber, fat tissue, dermis, etc.) but also blood cells (macrophages, leukocytes, etc.). These are tissues that have the same embryonic origin. However, not all scientific authorities consider blood or lymph as connective tissue. The breast is an organ that contains mainly connective tissue.

[8]  For a 50-year-old woman with small breast cancer, the absolute benefit expected from radiotherapy is a reduction in breast cancer mortality of about 2-5%. For non-smokers, the absolute risk of developing lung cancer or cardiotoxic risk 10 years after breast cancer radiation therapy appears to be less than 1%. For smokers, the absolute risk is 4%.

Estimation des risques de la radiothérapie du cancer du sein : Evidence From Modern Radiation Doses to the Lungs and Heart and From Previous Randomized Trials, Taylor C. et al, J Clin Oncol. 2017 20 mai ; 35(15):1641-1649).

This lack of benefit on overall mortality is also noted in a report in French of 2015, Evaluation of intraoperative radiotherapy in breast cancer, October 2015, Note de cadrage, HAS, October 2015.

[9] The COPERNIC project:

[10] SAS NEOLYS ( ) was founded by the founders Gilles Devillers, Nicolas Foray, Julien Gillet-Daubin. It is based on the 15 years of work of the Radiobiology Group of UMR 1052 INSERM.

[11] This rapid test was recently accepted in Int J radiat Biol Oncol Phys

[12] Apoptosis is one of the forms of cell death by self-destruction (a kind of cell suicide), genetically programmed.

 [13] A prospective study consists of comparing the occurrence of a pathology in groups defined according to their exposure to a factor presumed to be responsible for the pathology.

[14] SAS NOVAGRAY is directed by Clémence Franc who co-founded NOVAGRAY with Prof. David Azria in 2015.  Pr AZRIA is head of the Oncologic Radiotherapy Department at the Cancer Institute of Montpellier (ICM) and radiobiology project leader in the team “Immunociblage et radiobiologie en oncologie” directed by André Pélerin at the head of the INSERM U1194 unit at the Cancer Research Institute of Montpellier (IRCM).

🛈 Nous sommes un collectif de professionnels de la santé, rassemblés en association. Nous agissons et fonctionnons sans publicité, sans conflit d’intérêt, sans subvention. Merci de soutenir notre action sur HelloAsso.
🛈 We are an French non-profit organization of health care professionals. We act our activity without advertising, conflict of interest, subsidies. Thank you to support our activity on HelloAsso.
Scroll to Top