Method for prediction of prognosis by human papillomavirus DNA integration pattern

A method for predicting the prognosis of and treating a patient with locally advanced cervical cancer within stage IB to stage IVB who received radiotherapy, the method comprising the steps of:

• (A) classifying the patient into group A or group B,
• (B) predicting the prognosis of cervical cancer, wherein the patient classified into group A has a good prognosis, and wherein group B has a hazard ratio of 1.4 to 7.44 when the hazard ratio regarding the survival rate of group A is set at 1, wherein the hazard ratio is measured using a multivariable cox proportional hazards model, and
• (C) administering radiation and/or an anti-cancer agent to the patient, wherein the level of radiation and/or anti-cancer agent administered to the patient is reduced if the patient is classified into group A relative to the level to be administered if the patient is classified into group B;
• wherein the step (A) comprises:
• (I) obtaining a cervical tumor tissue from the patient with locally advanced cervical cancer within stage IB to stage IVB who received radiotherapy, wherein the cervical tumor tissue contains tumor cells and wherein the tumor cells are cells in the tumor tissue that contain human papillomavirus (HPV) DNA,
• (II) determining the HPV DNA pattern in tumor cells of the tumor tissue by in situ hybridization (ISH),
• (III) determining the percentage of the tumor cells with an episome pattern in the tumor tissue,
• (IV) classifying the patient into group A if the tumor tissue has an episome pattern in 80% to 100% of the tumor cells, and
• (V) classifying the patient into group B if the tumor tissue does not have an episome pattern in 80% to 100% of the tumor cells.