Biomarkers for diagnosis and prognosis of lung cancer
Provided herein are methods for non-invasively diagnosing and/or prognosing a lung cancer and for determining the efficacy of a therapeutic treatment regimen for the lung cancer. Expression levels of at least two small non-coding RNAs, for example, microRNAs and small nucleolar RNAs, are measured and used to calculate an area under the curve (AUC) that provides a probability of lung cancer in the subject. The smoking history of the subject and, if present, the size of pulmonary nodules may be incorporated into the calculation.
1. A method for diagnosing and treating a non-small cell lung cancer in a human subject, comprising:
a) obtaining a sputum sample from a human subject having a history of smoking;
b) obtaining a control sputum sample from a cancer free control human subject having a history of smoking;
c) measuring an expression level of each of MIR486-1 and MIR126 genes in the sputum sample and in the control sputum sample;
d) applying a logistic regression model:
Pr ( y= 1,cancer| U 1 )=exp( U 1 )/1+exp( U 1 ), where,
U 1 =−0.302+0.175×log 2 (MIR486-1)+0.336×log 2 (MIR-126);
e) diagnosing non-small cell lung cancer in the human subject when the Pr (y=1) value is greater than 0.298; and
f) administering a chemotherapeutic drug to the human subject diagnosed with non-small cell lung cancer.
2. The method of claim 1 , wherein step d) further comprises adding values representative of pack-years of smoking for the human subject and for the human control subject to the logistic regression model.
3. The method of claim 1 , wherein step d) further comprises adding values representative of a size of pulmonary nodules in the human subject and in the human control subject to the logistic regression model.
4. The method of claim 1 , further comprising in the human subject diagnosed with non-small cell lung cancer:
repeating steps a) to d) at intervals during and after administering the chemotherapeutic drug for the non-small cell lung cancer; and
comparing the Pr (y=1) value calculated at each interval with the initial Pr (y=1) value; wherein a comparison at each interval is an indicator of a therapeutic effectiveness of the chemotherapy drug.
5. A non-invasive method for early detection and treatment of a non-small cell lung cancer in a human subject, comprising:
a) obtaining a sputum sample from a human subject with a history of smoking and pulmonary nodules;
b) obtaining a control sputum sample from a cancer free control human subject with a history of smoking and pulmonary nodules;
c) measuring expression levels of MIR486-1 and MIR126 genes in the sputum sample and in the control sputum sample;
d) applying a logistic regression model:
Pr ( y= 1,cancer| U 3 )=exp( U 3 )/1+exp( U 3 ), where,
U 3 =−6.302+0.484×log 2 (MIR486-1)+5.042×CT+log 2 SP (smoking-pack-years);
e) diagnosing the non-small cell lung cancer in the human subject when the Pr (y=1) value is greater than 0.298; and
f) administering a chemotherapeutic drug to the human subject diagnosed with non-small cell lung cancer.