ESTABLISHING THE PRESENCE OF THE T(15-17) IN SUSPECTED ACUTE PROMYELOCYTIC LEUKEMIA - CYTOGENETIC, MOLECULAR AND PML IMMUNOFLUORESCENCE ASSESSMENT OF PATIENTS ENTERED INTO THE MRC ATRA TRIAL

Detection of the t(15;17) or its molecular consequence, the PML-RAR al pha rearrangement, is critical for meaningful analysis of clinical tri als involving patients with suspected acute promyelocytic leukaemia (A PL), Its presence remains the best predictor of a favourable response to retinoids, such as ATRA, which in combination with chemotherapy con fer significant improvements in disease-free survival. We have evaluat ed the relative efficacy of RT-PCR, cytogenetics and PML immunofluores cence staining to identify the existence of the translocation in 100 p atients entered into the Medical Research Council (M.R.C.) ATRA trial. RT-PCR successfully identified PML-RAR alpha rearrangements in 93/100 patients, including 65 where only peripheral blood or post-induction marrow samples were available for analysis and in 12 patients in whom cytogenetic assessment failed to demonstrate t(15;17) due to poor-qual ity metaphases (10/12) or as a reflection of cryptic PML-RAR alpha rea rrangements (2/12). Parallel employment of the RAR alpha-PML assay con firmed expression of del(17q)-derived transcripts in 81% and permitted determination of the PML breakpoint (a potential independent prognost ic variable) in all 93 cases. Sequencing of RT-PCR products derived fr om 50 patients with 3' PML breakpoints revealed five bcr 2 cases, incl uding a novel exon 5 breakpoint. 35/81 (43%) patients with cytogenetic evidence of t(15;17) possessed additional karyotypic abnormalities. I n four patients with available buffy coat smears, lack of cytogenetic or molecular evidence of the t(15;17) was confirmed by a wild-type PML immunofluorescence nuclear staining pattern, in contrast to the chara cteristic microparticulate distribution detected in 14 patients with R T-PCR evidence of the rearrangement, However, although PML immunofluor escence staining is suitable for rapid determination of patients likel y to benefit from ATRA, this approach does not obviate the need for cy togenetic and RT-PCR analysis of all patients entered into APL clinica l trials, because both techniques provide additional information which may prone to be of independent prognostic significance.