Letters to the Editor
176 the absolute difference in gene expression level of IGF-1R among patients with different TC classes and D13 status is small, this can be accounted for by the log-transformation and normalization of gene expression data. Furthermore, and given the aforementioned considerations, there may be biologic relevance to such differences, particularly for cases with higher levels of expression. Our results fail to confirm the prognostic importance of IGF-1R and IGF-1 suggested by other investigators.7,9 This may be due to differences in the study population, the relatively shorter follow-up in our cohort or the sample size, and the limited number of events. However, it is important to note that powerful prognostic factors like t(4;14), t(14;16), D13 (Figure 2c and d) and plasma cell labeling index (data not shown) already demonstrate statistically significant prognostic impact in our cohort despite the relatively short follow-up, suggesting that IGF-1 and IGF-1R expression levels are unlikely to have important prognostic impact. In conclusion, our study suggests there may be differential requirement for IGF-1 by the different genetic subtypes of MM as highlighted by the differential IGF-1R expression, particularly among patients with the more aggressive variants of MM. Therefore, when identifying novel targets, it is important to consider the context of genetic heterogeneity. The potential greater activity of anti-IGF-1R strategies is promising in that it addresses an unmet need, better and more durable treatment for patients with high risk MM such as those with t(4;14)(p16;q32) or t(14;16)(q32;q23).
Acknowledgements R Fonseca is a Clinical Investigator of the Damon Runyon Cancer Research Fund. This work is supported by Pfizer Inc., and Grants R01 CA83724-01, SPORE P50 CA100707-01and P01 CA62242 from the National Cancer Institute. WJ Chng is funded by an International Fellowship from the Agency for Science, Technology and Research (AnSTAR), Singapore. 1
WJ Chng1, A Gualberto2, R Fonseca1 Division of Hematology-Oncology, Mayo Clinic Scottsdale, Scottsdale, AZ, USA; and
2
Pfizer Global Research and Development, New London, CT, USA E-mail:
[email protected]
References 1 Georgii-Hemming P, Wiklund HJ, Ljunggren O, Nilsson K. Insulinlike growth factor I is a growth and survival factor in human multiple myeloma cell lines. Blood 1996; 88: 2250–2258. 2 Pollak MN, Schernhammer ES, Hankinson SE. Insulin-like growth factors and neoplasia. Nat Rev Cancer 2004; 4: 505–518. 3 Mitsiades CS, Mitsiades NS, McMullan CJ, Poulaki V, Shringarpure R, Akiyama M et al. Inhibition of the insulin-like growth factor receptor-1 tyrosine kinase activity as a therapeutic strategy for multiple myeloma, other hematologic malignancies, and solid tumors. Cancer Cell 2004; 5: 221–230. 4 Bergsagel PL, Kuehl WM, Zhan F, Sawyer J, Barlogie B, Shaughnessy Jr J. Cyclin D dysregulation: an early and unifying pathogenic event in multiple myeloma. Blood 2005; 106: 296–303. 5 Fonseca R, Debes-Marun CS, Picken EB, Dewald GW, Bryant SC, Winkler JM et al. The recurrent IgH translocations are highly associated with nonhyperdiploid variant multiple myeloma. Blood 2003; 102: 2562–2567. 6 Tai YT, Podar K, Catley L, Tseng YH, Akiyama M, Shringarpure R et al. Insulin-like growth factor-1 induces adhesion and migration in human multiple myeloma cells via activation of beta1-integrin and phosphatidylinositol 30 -kinase/AKT signaling. Cancer Res 2003; 63: 5850–5858. 7 Bataille R, Robillard N, Avet-Loiseau H, Harousseau JL, Moreau P. CD221 (IGF-1R) is aberrantly expressed in multiple myeloma, in relation to disease severity. Haematologica 2005; 90: 706–707. 8 Shaughnessy J, Jacobson J, Sawyer J, McCoy J, Fassas A, Zhan F et al. Continuous absence of metaphase-defined cytogenetic abnormalities, especially of chromosome 13 and hypodiploidy, ensures long-term survival in multiple myeloma treated with Total Therapy I: interpretation in the context of global gene expression. Blood 2003; 101: 3849–3856. 9 Standal T, Borset M, Lenhoff S, Wisloff F, Stordal B, Sundan A et al. Serum insulinlike growth factor is not elevated in patients with multiple myeloma but is still a prognostic factor. Blood 2002; 100: 3925–3929.
Validation of immunoglobulin gene rearrangement detection by PCR using commercially available BIOMED-2 primers
Leukemia (2006) 20, 176–179. doi:10.1038/sj.leu.2404049; published online 24 November 2005
Polymerase chain reaction (PCR)-based assays are commonly used to detect clonal immunoglobulin (IG) gene rearrangements during the evaluation of lymphocyte infiltrates. These assays are based on the following principles: B-cell precursors rearrange two or more of their six IG genes (two each IGH, IGK, IGL) during development such that each cell has a unique sequence (of unique length) in the V(D)J region (see Figure 1, also Trainor et al1 Coad et al2 vanDongen et al3 for review). Using PCR, the V(D)J regions in DNA isolated from a cell population of interest can be amplified and the size spectrum of fragments analyzed using electrophoresis. Nonrearranged genes will not amplify Leukemia
because the distance between primers is too great. Amplification from a polyclonal population of cells with rearranged IG genes produces fragments of varying lengths while amplification from a monoclonal population produces IG fragments of identical length. PCR-based clinical assays for IG gene rearrangement evaluation have been limited by lack of standardization and poor clinical sensitivity.4 Poor sensitivity is generally due to the use of a limited number of consensus V region primers that fail to bind when target sequences deviate from the germ line sequence. This is usually caused by point mutations introduced during the process of somatic hyper mutation in the secondary follicle center. The sensitivity of the IG gene rearrangement assay can be improved by increasing the number of primers used and the number of genes analyzed, but this is technically challenging. Recently, an extensive set of primers for PCR-based
Letters to the Editor
177
Figure 1 IG Genes and BIOMED-2 Primer Coverage. A schematic of the three IG gene V(D)J regions following rearrangement. The binding coverage of BIOMED-2 primers is also shown in stylized form. V, D, J and C refer to variable, diversity, joining and constant segments of the IG gene, respectively. See van Dongen et al 3 for germ line IG gene structure, exact primer locations and primer sequences.
Table 1
Summary of clinical sensitivity and specificity results
Diagnosis*
Normal* Reactive Angioimmunoblastic T-cell lymphoma
IGH and IGK analyzed Fresh (liquid or frozen solid tissue) Positive (percent) 0/26 (0) 0/30 (0) 1/7 (14) 4/4 (100)
IGH only analyzed
Paraffin Positive (percent)
Fresh (liquid or frozen solid tissue) Positive (percent)
Positive (percent)
F 0/8 (0) 0/1 (0)
0/26 (0) 0/30 (0) 0/7 (0)
F 0/8 (0) 0/1 (0)
F
F
Precursor B-cell acute lymphoblastic leukemia* B-cell chronic lymphocytic leukemia Mantle cell lymphoma Follicular lymphoma Splenic marginal zone lymphoma Extranodal marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue Lymphoplasmacytic lymphoma Diffuse large B-cell lymphoma
23/23 (100)
10/10 (100)
22/23 (96)
8/10 (80)
12/14 29/29 11/11 2/2
5/5 11/11 6/6 1/1
12/14 19/29 11/11 1/2
4/5 6/11 3/6 0/1
Plasma cell or multiple myeloma*
19/20 (95)
Classical Hodgkin lymphoma Nodular lymphocyte predominant Hodgkin lymphoma
5/12 (42) 1/5 (20)
(86) (100) (100) (100)
6/6 (100) 18/22 (82)
(100) (100) (100) (100)
F 6/6 (100)
4/4 (100)
Paraffin
(86) (66) (100) (50)
(80) (55) (50) (0)
6/6 (100) 17/22 (77)
F 4/6 (67)
F
16/20 (80)
F
2/4 (50) 1/3 (33)
4/12 (33) 1/5 (20)
2/4 (50) 1/3 (33)
* denotes diagnostic categories containing only blood or bones marrow samples. The remaining categories contain only solid tissue samples, except for the B-CLL category in which 4 blood or bone marrow samples were included.
IG gene rearrangement analysis was developed by the European BIOMED-2 collaborative study group.3 The goal was to optimize sensitivity and specificity while retaining clinical practicality in an attempt to standardize clinical IG gene rearrangement testing. A total of 41 primers were developed to cover the vast majority of theoretical rearrangements in all
three IG genes and grouped into 8 tubes: IGH-28 primers in tubes A-E; IGK-10 primers in tubes A and B; IGL-3 primers in a single tube (see Figure 1). The primers were validated using fresh tissue including 15 reactive samples and 49 well characterized B-cell and plasma cell neoplasms, all containing clonal IG gene rearrangement by the Southern blot technique. The correspondLeukemia
Letters to the Editor
178 ing paraffin-embedded tissue for seven reactive and 25 neoplastic samples was also evaluated. The result of this preliminary validation indicated the new primers would be useful for clinical testing with the maximum clinical sensitivity reported to be 98% in fresh tissue and approximately 75% in paraffin-embedded tissue. Analytical sensitivity varied from o1 to 10% depending on the primer set and the composition of the background cells. The BIOMED-2 primers are now commercially available for clinical use but a recent search of the English language literature revealed essentially no expansion of the initial validation studies such that laboratories introducing the primers must rely on data from the relatively few samples tested exclusively by the laboratories of the BIOMED-2 group prior to commercial production. We have performed independent validation of BIOMED-2 primers for IG gene rearrangement analysis in anticipation of primer use for routine clinical laboratory testing. A total of 211 fresh samples (26 normal, 30 reactive, 155 neoplastic) were tested and 55 of the samples (eight reactive, 47 neoplastic) had a corresponding paraffin-embedded sample adequate for parallel testing (Table 1). All samples were collected at Mayo Clinic with IRB approval and each case was reviewed to confirm both the diagnosis and the presence of adequate tumor to avoid false negative results due to insufficient sample. Genomic DNA was extracted using standard clinical methods and amplified using BIOMED-2 primers and the manufacturer’s instructions.5 Fragment analysis was performed using capillary electrophoresis in an automated genetic analyzer. Representative tracings of results are shown in Figure 2. A sample was considered positive if a clearly predominant fragment was observed in one or more acceptable size range in any PCR reaction with the exception of peaks in the IGK A tube 120–160 bp size range (see below and Figure 3). Equivocal peaks (discreet peaks o1000 fluorescence units) were sometimes observed and considered negative for the purpose of this study. Clinical sensitivity and specificity data is summarized in Table 1. Sensitivity for non-Hodgkin neoplasms ranged from 82 to 100% depending on tumor and specimen type. The diagnostic categories with o100% sensitivity were plasma cell neoplasms (95%), mantle cell lymphoma (86%), and diffuse large B-cell lymphoma (82%), none of which commonly pose a diagnostic dilemma requiring molecular evaluation. The cause of less than 100% sensitivity in these categories is not entirely clear. Less than perfect assay precision could be a factor in some cases, particularly if there should happen to be the combination of positivity in a single size range and poor amplification. When formally tested, precision was adequate but not 100%, as there was some inter-run variation in peak heights. However, this did not interfere with the result status of any of these cases and all negative cases in the study were retested with identical negative results. Thus, we do not feel routine duplicate testing is warranted, although it may be helpful in cases interpreted as equivocal on initial testing. Of course it is essential that adequate and representative tissue be tested to ensure optimal sensitivity and despite our attempts to control for this variable, it could still be a contributing factor, particularly in the plasma cell neoplasms where patchy tissue involvement is common. Clinical sensitivity was not significantly different between neoplasms considered prefollicular (expected to have no somatic hyper mutation) and those considered follicular or postfollicular (expected to have somatic hyper mutation). This indicates that the approach of analyzing multiple genes with multiple primers largely overcomes the problem of primer incompatibility. Further support for this concept comes from the Leukemia
Figure 2 Representative Tracings of Results. Composite of capillary electrophoresis tracings from a negative sample (a) and positive sample (b) showing all primer tubes in the IGH (IGH A–E) and IGK (IGK A and B) tubes. Expected fragment size ranges (base pairs) are shown with bars as follows: IGH tube A: 310–360; IGH tube B: 250–295; IGH tube C: 100–170; IGH tube D: 110–290 and 390–420; IGH tube E: 110–130; IGK tube A: 120–160 and 190–210 and 260– 300; IGK tube B: 210–250 and 270–300 and 350–390. Amplified fragments outside of these ranges are considered nonspecific.
Figure 3 Non-specific Results From the IGK-A tube 120–160 bp size range. Representative tracings from a negative sample (a) and a positive sample (b) showing a prominent reactive peak in the negative tracing (arrow) that might mistakenly be interpreted as positive. This peak can be compared with a true positive peak in the positive sample (arrow) that is very discreet and is not associated with a polyclonal background.
surprising result of moderate clinical sensitivity (35%) using the primer sets to evaluate Hodgkin lymphoma. Although the neoplastic cells in both nodular lymphocyte predominant and classical Hodgkin lymphoma contain rearranged IG genes,6,7 they contain extensive somatic hypermutation and represent a
Letters to the Editor
179 small percentage of the total cells in most samples such that clinical laboratory clonality studies are usually negative. The BIOMED-2 primers showed roughly equivalent clinical sensitivity for fresh and paraffin-embedded tissue samples. In general, the number of positive peaks was fewer, and the height of positive peaks less, in the paraffin samples compared with the corresponding fresh samples. However, this pattern was reversed in a small number of samples. The apparent overall increase in clinical sensitivity using paraffin tissue in this study (Table 1) is likely due to the fact that many of the negative fresh samples had inadequate paraffin samples to evaluate due to the retrospective study design; many of the paraffin samples had been pretreated with B5, a heavy metal fixative that is used to improve cell histology, but inhibits PCR reactions. However, in our clinical experience with freshly-embedded, non-B5 fixed samples, the majority are adequate for analysis. Maximum sensitivity was obtained when both IGH and IGK were evaluated (Table 1). In all, 60% of positive samples showed positivity in more than one size range. Primers in IGH tubes D and E produced positivity in the fewest number of samples and were rarely the sole positive primer groups (D tube positive 24% and sole positive 3%; E tube positive 1% and sole positive 0%). Thus, the true contribution to sensitivity of the primers in the IGH D and E tubes will require more extensive evaluation. The IGL primers did not appear to add sensitivity as all neoplastic samples found negative following evaluation of IGH and IGK were tested using the IGL primers without additional positive samples being detected. The distribution of primer positivity is similar to that reported in the original validation studies.3 Clinical specificity was 100% with all 64 normal and reactive samples being negative. However, many of these samples showed a large reactive peak in the smallest size range of the IGK A tube primers, which could easily be misinterpreted as a clonal peak. After careful review of the entire data set, it was noted that true positive peaks in the range of concern were distinctive, originating from the baseline without associated polyclonal background, but reactive peaks were less defined and always associated with some degree of polyclonal background (Figure 3). Despite this distinction, extreme caution should be used when apparent true positivity is identified only in this size range. We feel that such isolated peaks should be considered equivocal, particularly in the absence of data from other diagnostic methods for correlation. Analytical sensitivity was at least 1% in a T-cell rich (thymus) background but was only 30% in a B-cell rich (tonsil)
background. In addition, the presence of a limited amount of IG DNA in the sample (limited template) could not be shown to produce false-positive peaks. In conclusion, the BIOMED-2 primers for PCR-based evaluation of IG gene rearrangements performed very well in this independent validation study and the study significantly expand published validation data for these primers. For most clinical applications, sensitivity should be essentially 100%, as the neoplasm groups showing o100% sensitivity rarely require IG gene rearrangement analysis for diagnosis. However, analysis of both IGH and IGK is required for optimal sensitivity. Clinical specificity was found to be 100%.
RF McClure, P Kaur, E Pagel, PD Ouillette, CE Holtegaard, CL Treptow and PJ Kurtin Laboratory Medicine and Pathology, Division of Hematopathology, Mayo Clinic, Rochester, MN USA E-mail:
[email protected]
References 1 Trainor KJ, Brisco MJ, Wan JH, Neoh S, Grist S, Morley AA. Gene rearrangement in B- and T-lymphoproliferative disease detected by the polymerase chain reaction. Blood 1991; 78: 192–196. 2 Coad JE, Olson DJ, Lander TA, McGlennen RC. Molecular assessment of clonality in lymphoproliferative disorders: I. Immunoglobulin gene rearrangements. Mol Diagn 1996; 1: 335–355. 3 vanDongen JJM, Langerak AW, Bruggemann M, Evans PAS, Hummel M, Lavender FL et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 concerted action BMH4-CT98-3936. Leukemia 2003; 17: 2257–2317. 4 Bagg A, Braziel RM, Arber DA, Bijwaard KE, Chu AY. Immunoglobulin heavy chain gene analysis in lymphomas. J Mol Diagn 2002; 4: 81–89. 5 InVivoScribet Technologies. IGH gene clonality, IGK gene clonality, IGL gene clonality assays – ABI fluorescence detection – instruction manuals. 2005, 1–15. 6 Braeuninger A, Kuppers R, Strickler JG, Wacker H-H, Rajewsky K, Hansmann M-L. Hodgkin and Reed-Steernberg cells in lymphocyte predominant Hodgkin disease represent clonal populations of germinal center-derived tumor B-cells. Proc Natl Acad Sci 1997; 94: 9337–9342. 7 Marafioti T, Hummel M, Foss H-D, Laumen H, Korbjuhn P, Anagnostopoulos I et al. Hodgkin and Reed-Sternberg cells represent an expansion of a single clone originating from a germinal center Bcell with functional immunoglobulin gene rearrangements but defective immunoglobulin transcription. Blood 2000; 95: 1443–1450.
MLL gene rearrangements have no direct impact on Ara-C sensitivity in infant acute lymphoblastic leukemia and childhood M4/M5 acute myeloid leukemia
Leukemia (2006) 20, 179–182. doi:10.1038/sj.leu.2404031; published online 24 November 2005
The antimetabolite cytosine arabinoside (Ara-C) is a deoxycytidine analog that is used as a therapeutic agent in many leukemia treatment regimens. In combination with anthracyclines, Ara-C is the most effective agent in the treatment of acute myeloid leukemia (AML). In the treatment of acute lymphoblastic leukemia (ALL), the use of Ara-C is limited. However, leukemic cells from infants (o1 year of age) with ALL,
which are resistant to several chemotherapeutic drugs, are in vitro more sensitive to Ara-C as compared to older children with ALL.1,2 This observation suggested that infant ALL might resemble a subclass of childhood ALL, which may benefit from intensified treatment with Ara-C to improve the dismal prognosis for these patients who to date experience an event-free survival (EFS) of B35%. Therefore, in 1999, a novel collaborative treatment protocol (INTERFANT-99) was designed with intensified use of Ara-C, in order to provide a more specific treatment to infant ALL patients. Leukemia
