American Journal of Hematology 53:169-174 (1996)

Southern Technique and Cytogenetics Are

Complementary and Must Be Used Together in the
Evaluation of Phl M-BCR Positive Chronic Myeloid
Leukemia (CML) Patients Treated With Alpha Interferon
(IFN-ALPHA)
Juan Luis Steegmann, Maria Jose Requena, Luis Felipe Casado, Monica Pico,
Maria Jesus Peiiarrubia, Maria Teresa Ferro, Monica Resino, and
Jose Maria Fernandez-Raiiada
Genetics Department, Hospital Ramon Y Cajal (M.T.F., M.R.), and Hematology Department, Hospital d e la Princesa,
(J.L.S., M.J.R., L.F.C., M.P., M.J.P., J.M.F.-R.) Madrid, Spain

Cytogeneticanalysis is the gold standard for the follow-up of CML patients. The sensitivity
of cytogenetics is fairly similar to that of Southern detection of M-BCR rearrangement
(5%); this last technique has the potential advantage of being independent of cell division
and yield of metaphases. IFN alpha treatment can induce lack of growth of hemopoietic
precursors and poor yield of metaphases has been observed. For this reason we decided
to study the grade of concordance and complementarity between analysis of karyotype
and detection of M-BCR rearrangement of Southern blot. We studied 43 Phl positive, M-
BCR positive pre-BMT CML patients (48 samples) treated with IFN alpha 2a. Karyotype
was done on bone marrow cells by direct method, culture, and banding. Southern tech-
nique was performed onto DNA from peripheral blood leukocytes treated with Bglll (and
Xbal if necessary) and hybridized with the universal probe (Phllbcr-3, Transprobe 1)
labelled with dCTP32.
A highly significant association between both tests was obtained. Of 48 samples ana-
lyzed, 34 were evaluable by both methods and 28 gave the same result for both tests.
The concordance between the tests was good (kappa index: 0.63). Of total samples 27.1%
was not evaluable by cytogenetics; this figure was 31.2% in samples from patients who
were previously in complete cytogenetic response. All of the specimens not evaluable
by karyotyping were evaluable by Southern. One sample was not analyzable by Southern
but it was evaluable by cytogenetic analysis. The information obtained by Southern
technique was clinically relevant, and decisions were made according to its results.
We conclude that both tests show a significant association and a good concordance,
although they are not interchangeable.Cytogeneticand molecular studies are complemen-
tary and must be employed together in CML patients treated with alpha-interferon.
0 1996 Wiley-Liss, Inc.

Key words: chronic myelogenous leukemia, interferon alpha, cytogenetics, M-BCR,

Southern

INTRODUCTION 20% of major (complete and partial) cytogenetic re-

sponses can be obtained [3]. Obtaining a sustained major
Chronic myeloid leukemia is a clonal myeloprolifera-
genetic response in IFN-treated CML patients is crucial
tive disorder of the primitive haematopoietic stem cell
[l]. It displays a cytogenetic hallmark in more than 95%
of cases: the Phl chromosome that results from the recip-
Received for publication December 11, 1995; accepted June 18, 1996.
rocal translocation t9:22(q34;qll) which at a molecular
level represents the bcr-abl rearrangement 121. With IFN Address reprint requests to Juan Luis Steegmann, Hospital de la Pnn-
alpha treatment 70% of haematologic responses and 15- cesa, CI Diego de Leon, 62, Madrid 28006, Spain.
0 1996 Wiley-Liss, Inc.
170 Steegmann et al.
because of its prognostic value. A longer survival for
patients with IFN-induced major genetic response has
been suggested [4] and patients with complete genetic
response have an actuarial probability of survival without
genetic relapse in excess of 90% [5]. Certifying a com-
plete genetic response is thus the major issue in the fol-
lowup of CML patients treated with alpha-IFN.
Karyotype and molecular techniques (PCR and South-
em blot) can be employed to determine the grade of
leukemic Phl clone suppression in IFN treated CML
patients [6,7]. Although karyotype is the gold standard for
monitoring IFN treated CML patients [3] this technique
requires a bone marrow sample; besides it needs the
growth of an adequate number of metaphases [8]. South-
ern detection of M-BCR rearrangement has a sensitivity
similar to karyotyping (5-1076) [6] and is a laborious
and long technique but it has the advantage of not requir-
ing bone marrow (it can be performed on peripheral blood
leukocytes) and of being independent of cell division
[ 6 ] .This last point is of importance because IFN treated
patients may show lack of growth of metaphases in karyo-
typing [9]. After having detected this problem in some
of our first patients, our group decided to study samples
from IFN alpha-treated CML patients in order to appraise Fig. 1. Southern blot analysis. Autoradiographs of South-
the concordance and complementarity between cytoge- ern blot hybridized with the phllbcr-3 probe. Lanes 2,5,6,
netic analysis and detection of the M-BCR region re- show normal Bglllfragments. Lanes 1,3,4,7representnormal
and rearranged DNA fragments.
arrangement by Southern blot.

metaphases; partial response (PGR) indicated 5-34% Ph

PATIENTS AND METHODS positive metaphases; a minimal cytogenetic response was
Patients defined as presence of the Ph chromosome between 35
and 95% of metaphases.
We have studied 43 patients diagnosed of chronic my- Southern blot was performed onto DNA extracted from
eloid leukemia Phl, M-BCR positive in first chronic peripheral blood leukocytes by standard method (phenol-
phase. All of them were treated with interferon alpha 2a chlorophorm extraction and precipitation with etanol)
(Roferon-A): 5.4 2 3 million U/d for a median time of [ll]. DNA was digested with the restriction enzymes
7 1 I days (range: 241,826). The number of samples ana- BgIII and Xbal if necessary (in less than 1% of cases
lyzed as 48, of which 34 were from patients in complete abnormal restriction fragments may not be visible after
hematologic remission and 16 were from patients in previ- digestion with BgIII), electrophoresed in 0.8% agarose
ous complete cytogenetic remission. We did not discon- gel, transferred to a nylon membrane (Gene Screen Plus,
tinue IFN treatment before obtaining bone marrow or Du Pont, Wilmington, DE) in 0.4 N NaOH and hybridized
blood specimens. Informed consent was obtained ac- with the universal probe (Phlhcr-3 Transprobe 1, Onco-
cording to local-ethic committee guidelines. gene Science, Manhasset, NY) labelled with dCTP32
[ 121. The technique was qualitative and densitometry of
Methods bands was not perfomled. An autorradiography sample
The karyotype analysis was carried out on bone marrow is shown in Figure 1.
cells extracted in heparin. Some cells were cultured in Statistical methods. For statistical consideration, a
RPMI1640 medium supplemented with 18% fetal bovine cytogenetic analysis was scored as non-evaluable when
serum for 2 hr; then colchicine was added followed by no metaphases were obtained, as positive when Phl posi-
KCL 0.075 M and afterwards they were fixed with Carnoy tive metaphases were observed, and negative when no
and observed (direct method). The rest of the cells were Phl metaphases could be visualized; the target number
cultured for 24 hr and G banding technique was accom- of metaphases to be counted was 20. Southern results
plished [lo]. were classified as non-evaluable when no bands were
Cytogenetic response to IFN was classified according present or the intensity of them was too faint to conclude,
to Talpaz et al. [3]: CGR was defined as no Ph positive as positive in the presence of additional bands, and as
 Southern and Cytogenetics in CML 171

TABLE 1. Previous Cytogenetic Response to IFN Classified TABLE 111. Association and Concordance of Southern
According to Talpaz et al. [3] and KarVOtvDe.

Cytogenetic response (CR) N Karvotvue

16 Phl positive Phl negative

Complete
Partial 2
Southern
Minor 11
Positive 17 2
No CR 17 4
Negative 11
Not assessed 2
*Chi square: 13.59 P < 0.0002; kappa index: 0.66.

TABLE IV. Evaluable Results bv Each Techniaue*

TABLE II. Number of Metaphases Obtained From Bone
Karvotvue
Marrow Aspiration
Evaluable Not evaluable
No. of metaphases N
Southern
Zero 13 Evaluable 34/10 1315
1-10 1 Not evaluable 111 010
11-20 12
>20 15 *Total samples (n = 48, bold type) and samples from patients with previous
Not stated I CGR (n = 16, italics).

was 0.63; this value shows a fairly good concordance

between the tests.
negative when only the germ line bands were observed.
The association was determined by the Chi2 test and the Complementarity
concordance was estimated by the kappa index. Of total samples 27.1% was not evaluable by karyo-
type; out of 16 samples from patients with previous CCR,
5 were not analyzable by karyotype. All of them could
be evaluated by Southern. One sample that could not
RESULTS be evaluated by Southern was evaluable by karyotyping
We analyzed 48 samples from 43 patients of chronic (Table IV).
phase CML treated with alpha 2a IFN (mean + SD: Clinical Relevance
5.4 -t 3 million U/d) for a median time of 711 days (range:
261,826). Out of 48 specimens, 34 were from patients The clinical value of the detection of M-BCR re-
in complete hematologic remission, 9 from patients with arrangement in the setting of non-evaluable cytogenetics
partial hematologic response, and 5 belonged to individu- is depicted in Table V, where the results of the Southern
als with no response. Table I shows the previous cytoge- technique are compared with the immediately previous
netic response the patients obtained before the problem and next cytogenetic response.
sample was obtained. It can be seen that 16 out of 48 Patient 22 was treated with alpha IFN after relapsing
samples were from patients who have had a previous post BMT. MBCR was rearranged at this moment (Fig.
CGR. The number of metaphases obtained is depicted in 1, lane 4). In this patient absence of MBCR rearrangement
Table 11; it is important to point out that no metaphases allows us to certify the complete response, sparing the
could be obtained in 13 cases. As a control, it must be need of alternative treatment and of marrow punctures
pointed out that in 15% of 120 samples from patients (Fig. 1, lane 5). This patient continues in IFN maintained
with several hematologic diseases not receiving IF", the CGR, 84 months after IFN was started and 69 months after
growth of metaphases could not be achieved. obtaining the CGR. Her evolution has been previously
described [13]. No bcr-abl transcript is detected by double
step-RT-PCR (data not shown) [14].
Concordance and Association Cytogenetic studies on patient 16 were extremely dis-
Of 48 samples analyzed, 34 were evaluable by both appointing due to frequent lack of metaphases, and we
methods and 28 gave the same result for both tests. We used Southern technique in order to monitor her complete
found a highly significant association between cytoge- response, which persists more than 70 months after its de-
netic analysis and Southern by means of the Chi 2 test tection.
(Chi2 = 13.59;P < 0.0002) (Table 111). The kappa index Patient 72 is a child who obtained a complete cytoge-
172 Steegmann et al.

TABLE V. Results Obtained by Southern (Central Column) in Samples Which Were

Coincidental With a Non-Evaluable Cytogenetic Exam (i.e., No Metaphases Were Obtained)*

Previous GR Next GR
Patient no. 1% Phl (no. met)] MBCR 1% Phl (no. met)]
6 90 (19) Rearranged 100 (NS)
5 66 (20) Rearranged 45 (20)
8 5 (18) Rearranged 10 (20)
15 Not evaluable Rearranged 85 (NS)
16 0 (40)” Not rearranged Not evaluable
16 Not evaluable Not rearranged 0 (25)
16 0 (25) Not rearranged 0 (20)
20 100 (25) Rearranged Not evaluated
22 0 (15) Not rearranged 0 (15)
35 100 (13) Rearranged Not evaluatedb
44 100 (NS) Rearranged 100 (22)
72 15 (27) Not rearranged Not evaluatedb
10 100 (12) Rearranged 100 (NS)

*The left column shows the previous result obtained by cytogenetics in the same patient, and the right column
shows the cytogenetic results of the next immediate exam, if available. NS, not stated.
this patient, three exams were not evaluable by cytogenetics at dates 12/91, 10192, 11/93.
bThese patients were submitted to BMT

netic response after IFN. IFN was stopped in order to tients because most patients with a complete cytogenetic
collect bone marrow, but Phl metaphases increased to response show presence of the abnormal transcript bcr-abl
15% during this period. IFN was restarted after the bone [ 181. Preliminary reports recently claim that quantitative
marrow harvest and a disappearance of MBCR re- PCR seems to have definite advantages over karyotyping,
arrangement was reinduced; at that moment an unrelated although difficulties in its standardization may hamper
bone marrow donor was found. BMT was performed its widespread use [ 191. Interphase-fluorescent in situ
while on CGR, and the patient is currently alive 18 months hybridization (iFISH) has the theoretical advantage of
after BMT. not needing the yield of cell divisions; however, recent
Patients 20 and 35 were patients with poor tolerance reports comparing metaphase FISH and interphase FISH
to IFN. Absence of complete response in MBCR re- in CML patients have shown that iFISH overestimates
arrangement contributed to the discontinuation of the drug the degree of cytogenetic response [20].
on these patients. Therefore, karyotyping and detection of MBCR re-
arrangement are the mainstays of the follow-up of CML
patients. Contrary to quantitative PCR and FISH, these
DISCUSSION
techniques are available to most medium-sized hematol-
Cytogenetic studies are the gold standard for monitor- ogy departments. To our knowledge, our study is the first
ing the treatment of Phl positive CML patients. Southern which is addressed to assess the concordance of these
detection of MBCR rearrangement in peripheral blood two techniques in IFN treated CML patients. Our results
leukocytes and bone marrow karyotype have been com- seem to indicate that interferon alpha treatment in CML
pared in patients with CML in chronic phase treated with patients is frequently associated with a poor yield of
chemotherapy. In these studies, the results of MBCR metaphases; this fact could reflect the delaying effect of
rearrangement analysis in peripheral blood leukocytes IFN on the cell cycle [21]; alternatively, it may result
showed a good concordance with those of bone marrow from the myelosuppression induced by IFN. In our series,
karyotype (kappa index = 0.65) [15]. 50% of samples had a poor marrow cellularity and in
Detection of disappearance of Phl chromosome or its half of them no metaphases were obtained.
molecular counterpart, the MBCR rearrangement, has an Although no change in overall bone marrow cellularity
important clinical impact in CML patients treated with was found in patients with solid tumors treated with IFN
IFN alpha, because complete cytogenetic responses seem alpha for a short period of time [22], several authors have
to be associated with longer duration of hematologic re- observed that “emptiness” of bone marrow samples is a
sponses [3] and better survival [4]. Moreover, patients frequent finding in IFN-treated CML patients [23]. My-
in complete cytogenetic remission could be eligible for elosuppression is a rather frequent secondary effect of
marrow collection and, eventually, for autologous trans- IFN alpha in CML and even aplasia has been described
plantation [16,17]. in IFN alpha treated Phl positive CML patients that had
Qualitative PCR seems of limited utility in these pa- previously received alkylating agent [24]. A poor yield
 Southern and Cytogenetics in CML 173

of metaphases could lower the predictive power of the REFERENCES

technique [8]; the absence of growth of metaphases may 1. Clarkson B, Strife A: Linkage of proliferative and maturational abnor-
even indicate a new bone marrow aspiration. malities in chronic myelogenous leukemia and relevance to treatment.
In our study we found that Southern and karyotype Leukemia 7:1683, 1993.
analysis show a very significant association. The kappa 2. Rowley JD: A new consistent chromosomal abnormality in chronic
index greater than 0.6 demonstrates that the concordance myelogenous leukemia identified by quinacrine fluorescence and
Giemsa staining. Nature 243:290, 1973.
between the tests is good. However, Southern detection
3. Talpaz M, Kantarjian H, Kurzrock R, et al: Interferon-alpha produces
of MBCR rearrangement cannot replace cytogenetics in sustained cytogenetic responses in chronic myelogenous leukemia.
monitoring these patients. Besides, karyotyping permits Philadelphia chromosome-positive patients. Ann Intern Med 114:
the demonstration of other chromosomic anomalies that 532, 1991.
may be of prognostic significance [25], and gives a quanti- 4. Italian Cooperative Study Group on Chronic Myeloid Leukemia: Inter-
feron alfa-2a as compared with conventional chemotherapy for the
tative estimate of IFN effect; this makes an advantage over
treatment of chronic myeloid leukemia. New Engl J Med 330:820,1994.
the qualitative Southern technique we have employed. 5. Giralt S, Kantarjian H, Talpaz M: The natural history of chronic myelog-
Quantitative analysis of MBCR rearrangement has been enous leukemia in the interferon era. Semin Hematol 32:152, 1995.
thoroughly standardized by Ayscue et al. and Grossman 6. Morgan GJ, Wiedemann JM: The clinical application of molecular
et al. who applied this technique to monitor the IFN alpha techniques in Philadelphia-positive leukemia. Br J Haematol 80:1,
1992.
treatment in three and six patients, respectively [26,27].
7. Yoffe G, Blick M, Kantarjian H, et al: Molecular analysis of interferon-
Few samples non-evaluable by Southern are expected induced suppression of Philadelphia chromosome in patients with
(for instance, only one out of 48, in our series), and we chronic myeloid leukemia. Blood 69:961, 1987.
can anticipate that quantitative Southern may be a most 8. Hook EB: Exclusion of chromosomal mosaicism: Tables of 9O%, 95%,
valuable tool in the monitoring of these patients. It is and 99% confidence limits and comments on use. Am J Hum Genet
29:94, 1977.
conceivable that karyotype and MBCR analysis may re-
9. Higano CS, Raskind WH, Singer JW. Use of alpha interferon for the
flect different biological phenomena in CML patients treatment of relapse of chronic myelogenous leukemia in chronic phase
treated with IFN alpha, having potentially different popu- after allogeneic bone marrow transplantation. Blood 80: 1437, 1992.
lations as targets; dividing cells and total cells, respec- 10. Ferro MT, del Potro E, Krsnik I, et al: Unusual translocations and
tively. Further studies are necessary to clarify this point, other changes in acute leukemia. Cancer Genet Cytogenet 54: 163,1991.
11. Sambrook J, Fritsch EF, Maniatis T Analysis and Cloning and Eukaryo-
and its clinical relevance. Recently, Verschraegen et al.
tic Genomic DNA. Isolation of DNA From Mammalian Cells: Protocol
have applied a quantitative Southern technique in a very 1. Anonymous Molecular Cloning. A Laboratory Manual. Ed. 2. New
large group of patients, and their results showed an excel- York: Cold Spring Harbor Laboratory Press, 1989, p 16.
lent correlation between the two techniques. This study 12. Feinberg AP, Vogelstein B: A technique for radiolabeling DNA restric-
included only samples evaluable by cytogenetics (i.e., it tion endonuclease fragments to high specificity activity. Anal Biochem
132:6, 1983.
did not include samples without yield of metaphases) [28].
13. Steegmann JL, Perez M, Vazquez L, et al: Interferon alpha treatment
Conversely, our study was addressed to ascertain the of accelerated-phase chronic myeloid leukemia in relapse after bone
degree of complementarity, that is, if Southern would marrow transplantation: A case with complete cytogenetic and molecu-
allow to evaluate samples which were not evaluable by lar remission. Bone Marrow Transplant. 7:65, 1991.
cytogenetics, as a result of poor yield of metaphases. We 14. Steegmann JL, Diez JL: El interfer6n alfa y las infusiones de linfocitos
de donante en las recaidas de la leucemia mieloide cr6nica post-TMO.
have found that 27.1% of total samples were not evaluable
Sangre 40:229, 1995.
by cytogenetics, but they were successfully analyzed by 15. Blennerhasset GT, Furth ME, Anderson A, et al: Clinical evaluation
Southern. This technique has been very useful and clini- of a DNA probe assay for the Philadelphia (Phl) translocation in
cally relevant in patients with previous complete genetic chronic myeloid leukemia. Leukemia 2:648, 1988.
responses; in these patients karyotype was not evaluable 16. Hoyle C, Gray R, Goldman J: Autografting for patients with CML in
chronic phase: An update. Br J Haematol 86:76, 1994.
in 5 out of 16 samples, whereas Southern was informative
17. Simonsson B, Oberg G, Killander A, et al: Intensive treatment in order
in all of them. This information allowed us to take clini- to minimize the Ph-positive clone in chronic myelogenic leukemia.
cally important decisions, such as delaying donor lympho- Stem Cells Dayt 11(Suppl 3):73, 1993.
cyte infusions (IDN 22) or continuing with IFN (IDN 16) 18 Dhingra K, Kurzrock R, Kantarjian H, et al: Minimal residual disease
(see Results). in interferon-treated chronic myelogenous leukemia: Results and pit-
falls of analysis based on polymerase chain reaction. Leukemia
We conclude that Southern and cytogenetics are com-
6:754. 1992.
plementary and must be used together in the follow-up 19. Cross NC, Feng L, Chase A, et al: Competitive polymerase chain
of IFN alpha treated chronic phase CML patients. This reaction to estimate the number of BCR-ABL transcripts in chronic
seems to be especially true in patients with previous CGR; myeloid leukemia patients after bone marrow transplantation. Blood
in these patients a poor yield of metaphases is rather 82: 1929, 1993.
20. Kantarjian HM, Talpaz M, Hester J, et al: Collection of peripheral-
common, and, if the result of MBCR rearrangement is blood diploid cells from chronic myelogenous leukemia patients early
negative, new and disturbing bone marrow sampling in the recovery phase from myelosuppression induced by intensive
could be spared, and clinical decisions could be taken on dose chemotherapy. J Clin Oncol 13353, 1995.
its basis. 21. De Maeyer E, De Maeyer-Guignard J: The effects of interferons on
174 Steegmann et al.
cell growth and division. In De Maeyer E (ed): ‘‘Interferons and Other Chromosomes in Phl -positive chronic granulocytic leukemia. Br J
Regulatory Cytokines.” New York: John Wiley & Sons, 1988, p 134. Haematol 39:305, 1978.
22. Emstoff MS, Kirkwood JM: Changes in the bone marrow of cancer 26. Ayscue LH, Ross DW, Ozer H, et al: B d a b l recombinant DNA analysis
patients treated with recombinant interferon alpha-2. Am J Med versus karyotype in the diagnosis and therapeutic monitoring of Chronic
76593, 1984. Myeloid Leukemia. Am J Clin Path01 94:404, 1990.
23. Wandl UB, Butzler R, Niederle N, et al: Bcr-abl-positive and -negative 27. Grossman A, Silver RT, Szatroswki l”,et a]: Densitometric analysis
clonogenic cells in CML patients undergoing long-term interferon treat- of Southern blot autoradiographs and its application to monitoring
ment. Leukemia 8:776, 1994. patients with chronic myeloid leukemia. Leukemia 5:540, 1991.
24. TalpazM, Kantarjian H, Kurzrock R, et a]: Bone marrow hypoplasia and 28. Verschraegen CF, Talpaz M, Hirsch-Ginsberg CF, et al: Quantification
aplasia complicating interferon therapy for chronic myeloid leukemia. of the breakpoint cluster region rearrangement for clinical monitoring
Cancer 69:410, 1991. in Philadelphia chromosome-positive chronic myeloid leukemia. Blood
25. First International Workshop on chromosomes in Leukemia 1977: 85:2705, 1995.