Methods for predicting cancer drug responsiveness

A method for predicting a patient's sensitivity to a cancer treatment, comprising:

• (a) contacting a sample with an antibody or antibody format that recognizes a heterodimer comprising two B-cell lymphoma 2 (BCL-2) proteins, wherein the antibody or antibody format comprises:
• (i) a heavy chain variable region comprising heavy chain CDR1, CDR2, and CDR3 sequences, wherein the heavy chain CDR1 sequence is GHTFTEHYIN (SEQ ID NO:
• 1), the heavy chain CDR2 sequence is WIFPGSGSTYYNEKFKG (SEQ ID NO:
• 2);
• and the heavy chain CDR3 sequence is SYSNFWFAY (SEQ ID NO:
• 3);
• and
• (ii) a light chain variable region comprising light chain CDR1, CDR2, and CDR3 sequences, wherein the light chain CDR1 sequence is RASQSIGTSIH (SEQ ID NO:
• 4), the light chain CDR2 sequence is KYASESIS (SEQ ID NO:
• 5), and the light chain CDR3 sequence is QQSNSWPTT (SEQ ID NO:
• 6), the sample being a specimen from a solid tumor of the patient;
• (b) detecting a signal that indicates the amount of the heterodimer
• (c) determining a ratio of the amount of heterodimer present in the sample from step (b) to a reference value,
• wherein the reference value comprises the amount of one of the BCL-2 protein monomers of the heterodimer in the sample, and
• (d) determining a correlation between the ratio of the amount of heterodimer in the sample to the amount of one of the BCL-2 protein monomers of the heterodimer in the sample,
• the ratio being predictive of the patient's sensitivity to the cancer treatment.