Cytogenetic analysis has become an integral part of diagnosis and prognostication of acute myeloid (AML) and acute lymphoblastic leukemia (ALL). Consequently, CALGB mandates cytogenetic analyses for adult AML and ALL first-line treatment trials and many correlative studies. It is thus imperative that cytogenetic data are accurate. To this end, CALGB has performed central review of karyotypes (CRK) submitted by CALGB approved institutional cytogenetics laboratories since 1985. In this study, we have evaluated the role of CRK in ensuring that high quality cytogenetic data are available to CALGB researchers using two criteria: 1) the proportion of specimens deemed on CRK to be of sufficient vs insufficient quality to be included in the CALGB database (ie, accepted vs rejected according to the criteria published by

Byrd et al,
Blood
2002
;
100
:
4325
–4336
), and 2) among the accepted cases, the proportion of submitted karyotypes whose interpretation was changed during CRK. In our analyses, we excluded samples studied cytogenetically during complete remission, because they differ from pretreatment and relapse samples in that they rarely contain leukemic cells and usually are karyotypically normal. Overall, 18% of AML and 35% of ALL karyotypes submitted were not accepted (Table). The most common reason for rejecting a karyotype on CRK was inadequate banding quality, which accounted for rejection in 41% of AML and 34% of ALL inadequate cases.

DiseaseAll Specimens1986 CRK2001-2003 CRKP-value*
* A comparison of the rejection rates between the 1986 and 2001–2003 CRKs using Chi-square test; † % of the accepted specimens 
AML No. of specimens reviewed 4169 269 795 <0.0001 
 % rejected on CRK 18% 30% 12%  
 % revised on CRK†   26%  
ALL No. of specimens reviewed 996 92 65 0.004 
 % rejected on CRK 35% 49% 26%  
 % revised on CRK†   25%  
DiseaseAll Specimens1986 CRK2001-2003 CRKP-value*
* A comparison of the rejection rates between the 1986 and 2001–2003 CRKs using Chi-square test; † % of the accepted specimens 
AML No. of specimens reviewed 4169 269 795 <0.0001 
 % rejected on CRK 18% 30% 12%  
 % revised on CRK†   26%  
ALL No. of specimens reviewed 996 92 65 0.004 
 % rejected on CRK 35% 49% 26%  
 % revised on CRK†   25%  
View Large

The quality of the submitted karyotypes, measured by the proportion of rejected cases, has improved significantly in both AML and ALL since 1986. However, CRK in 2001–2003 still found 12% of AML and 26% of ALL samples inadequate. Among karyotypes deemed adequate, we analyzed in detail revisions made during the 2001–2003 CRK. Changes in karyotype interpretation were made in 26% of AML and 25% of ALL cases. The revisions included identification or reinterpretation, other than reassignment of breakpoints, of the chromosome abnormality (seen in 52% of samples with karyotype errors), a misidentified or upside down chromosome(s) (34%), reassignment of breakpoints in structural aberrations recognized by the submitting laboratory (28%) and correction of errors in the ISCN (1995) nomenclature (15%). Examples of clinically relevant changes in karyotype interpretation included revisions of a normal karyotype to an abnormal one that harbored inv(3)(q21q26), t(9;11)(p22;q23), t(11;19)(q23;p13.1), or inv(16)(p13q22), change from del(11)(q23) to t(6;11)(q27;q23), etc. Overall, 35% of AML and 45% of ALL samples submitted were either rejected or revised on CRK. We conclude that although we observed an improvement in quality of cytogenetic analyses over time, central review of karyotypes still plays a vital role in ensuring the success of the clinical trials and correlative studies conducted by cooperative groups.

Topics:
acute lymphocytic leukemia, cancer and leukemia group b, chromosomal disorder, clinical trials cooperative groups, complete remission, cytogenetic analysis, cytogenetics, karyotype determination procedure, leukemia, acute, leukemic cells

Author notes

Corresponding author

2005, The American Society of Hematology
2004
Sign in via your Institution