Create a base based on artificial intelligence that is capable of predict that patients with advanced cancer will respond to immunotherapy.
Developed by investigators from the Vall d’Hebron Institute of Oncology (VHIO), the study use more than one hundred date dates from images of solid tumors tomadas median a TAC to identify with a precision of up to 75% to those people whose susceptibility to respond to treatment.
This advance, which is taxed in the review Radiology, contribute to the story precision medicine, clave to be able to offer the most effective therapy for each person and increase their prognosis.
Immunotherapy is one of the best in cancer treatment most important of the last years. Unlike chemotherapy, which attacks tumor cells, this strategy works with the human immune system to destroy the tumor.
Geen hindernis nie, ya pesar de ser sumamente eficaz, not all patients respond to this type of pharmacy no tampoco padecen the same secondary effects y, at the moment, do not understand well the pore of their differences.
Although biomarkers are used, as well as determined mutations or genetic and epigenetic alterations that arrojan pistas about the tumor responds or not to the immunotherapy, son inadequate.
Investigator Raquel Pérez-López, of the VHIO Radiology Unit, led this study.
In this sense, the leader of the team of investigators, the radiologist Raquel Pérez, and his team he developedn model that applies a type of artificial intelligence to computed tomography (TAC) images obtenidas antes de that patients begin the treatment to realize a prediction about the illness.
“The way in which the radiologists analyze the images of the tumors of a patient are very subjective“Based on our experience”, explains Pérez, principal investigator of the radiology group – an emerging campus of radiology that extracts quantitative data from VHIO images.
“But its digital images, pixel composites and behind every pixel hooi information very valiosa that human beings can not perceive or analyze, but the program itself ”, said.
Junto a su equipo, Pérez analyzed the TAC of 236 solid tumors in advanced stages of 85 patients. It uses artificial intelligence of the machine learning type, and feeds it with more than a hundred quantitative values of each tumor and establish associations between the image and the molecular profiles related to the human resource.
In order to validate the model and test if it is actually capable of making reliable predictions, the researchers use other patient groups, one for cancer and the other for pulmonary disease, types of tumors for which immunotherapy is in place. those who have access to information regarding the response to treatment.
Scientists believe that the herramienta lograba identifies with a sensitivity of 85% patients with weight loss and 76% had pulmonary response to treatment.
“Sensitivity is moderately high in this test. a potential to better identify patients who can benefit from immunotherapy and, for that matter, in whom this treatment can give priority to others ”, considers this investigator.
La herramienta, geen invasive, permite, además, perform a tumor evolution follow-up, as well as the impact of treatment on the length of time. Asimismo, performs an evaluation of all tumors and only a small part, as can be seen when taking a biopsy, and this is a critical result, since suelen tumors are very heterogeneous and, on many occasions, the primary ones are small. with its metastasis.
“The TAC allows us to evaluate all the lesions of global and individual form,” said Pérez, explaining that there is now work to establish subgroups of tumors and to extract tumor characteristics that permit a mayor with the ability to respond.
The objective, the final, is power integrate the information that proportions the images of the lesions with the data obtained from the molecular characterization of the tumor, in order to increase the knowledge of cancer and the treatment of patients.
“Now when a person with cancer begins to receive a treatment of chemistry or radio, if you want to stay between 8 and 12 weeks to make a TAC and see if there are changes, even at the molecular level there are many things that can be identified beforehand. It is our goal: to use radio to see and quantify our antiquities, ”says Pérez.
The investigation was conducted by the Unit for the Investigation of Molecular Therapy of Cancer (UITM) – Foundation “la Caixa”, integrated by a multidisciplinary team of investigators, and has accounted for the impulse of this entity.
Source: La Vanguardia