Steps towards personalized therapy for lung cancer.
The National Cancer Institute (Int) of Milan is the leader of the I3LUNG international research project funded with 10 million euros by the EU: the goal is to identify, exploiting the potential of artificial intelligence, various possible biomarkers to make 'a patient size', and therefore more effective, immunotherapy against this malignancy.
Coordinates the project Arsela Prelaj, researcher of the Int.
The study is published in Clinical Lung Cancer.
Non-small cell lung cancer (NSCLC) is currently the most common type of lung cancer.
Immunotherapy is offered to patients with NSCLC who cannot access target therapies, but it is not effective in all cases.
Lung cancer is third in the ranking of the most diagnosed neoplasms - 40,000 cases a year in our country - and the leading cause of death from cancer in industrialized countries.
However, the possibility of administering the new therapies is subject to a precise molecular diagnosis and immunotherapy is remarkably effective in subjects who express high levels of a specific biomarker, the PD-L1 protein.
The problem, as pointed out by the researcher, is that this biomarker, approved because it is the best among those identified, is not perfect.
"In Italy, the current therapeutic approach provides that subjects with high PD-L1 can access immunotherapy alone, while those with low PD-L1 access immunotherapy associated with chemotherapy. However, there is a percentage of patients with low levels which responds equally well to immunotherapy alone", continues Prelaj.
Hence the researcher's idea that a better molecular profiling of the patient can be achieved not with one, but with more biomarkers, to be identified by analyzing the clinical data, in part already available and in part still to be collected, even with the use of AI, much more efficient than conventional methods in identifying correlations and salient information, which may have so far escaped the attention of researchers.
The current therapeutic approach provides that subjects with high PD-L1 can access immunotherapy alone, while those with low PD-L1 access immunotherapy associated with chemotherapy.
However, there is a percentage of patients with low levels who respond equally well to immunotherapy alone", continues Prelaj. Hence the researcher's idea that a better molecular profiling of the patient can be achieved not with one, but with more biomarkers, to be identified by analyzing clinical data, partly already available and partly still to be collected, even with the use of AI, much more efficient than conventional methods in identifying correlations and salient information, perhaps which have so far escaped the attention of researchers.
The current therapeutic approach provides that subjects with high PD-L1 can access immunotherapy alone, while those with low PD-L1 access immunotherapy associated with chemotherapy.
However, there is a percentage of patients with low levels who respond equally well to immunotherapy alone", continues Prelaj. Hence the researcher's idea that a better molecular profiling of the patient can be achieved not with one, but with more biomarkers, to be identified by analyzing clinical data, partly already available and partly still to be collected, even with the use of AI, much more efficient than conventional methods in identifying correlations and salient information, perhaps which have so far escaped the attention of researchers.