1. Art and Diseño

From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
Polymerase Chain Reaction Detection of the BCR-ABL Fusion Transcript
After Allogeneic Marrow Transplantation for Chronic Myeloid Leukemia:
Results and Implications in 346 Patients
By J.P. Radich, G. Gehly,t T. Gooley, E. Bryant, R.A. Clift, S . Collins, S. Edmands, J. Kirk, A. Lee, P. Kessler,
G. Schoch, C.D. Buckner, K.M. Sullivan, F.R. Appelbaum, and E.D. Thomas
We studied 346 patients after bone marrow transplantation
(BMT) for chronic myeloid leukemia (CML) for the presence
of the bcr-ab/ transcript detected by the polymerase chain
reaction (PCR) t o understand the frequency and implication
periphof a positive test. A total of634 samples of BM andlor
eral blood were obtained for PCR analysis between 3 and
192 months after BMT. A positive PCR test at3 months postBMT was not statistically significantly associated with an
increased risk of relapse compared with PCR-negative patients. However, a positive PCR assay at 6 months and beyond was highly associated with subsequent relapse. The
Kaplan-Meier estimate of relapse for patients testing PCRpositive at 6 t o 12 months was 4290 versus 3% for PCRnegative patients (PC .0001). The Kaplan-Meier estimate of
survival at 4 years for the PCR-positive patients was 74%
compared with 8340 for the PCR-negative group (P= .002).
Multivariable analysis indicated that a PCR-positive result at
6 to 12 months post-BMT, the type of
BMT donor (allogeneic
matched donor Y mismatched or unrelated), and the prescence of acute GVHD were independent risk factors for subsequent relapse. The relative risk(RR) for relapse for patients
PCR-positive at 6 t o 12 months post-BMT was 26.1(95%
confidence interval, 8.9 t o 76.1, P < .0001). The outcome of
long-term patients(>36 months post-BMT) who testedPCRpositive was much better, as 15 of 59 (2540)tested positive
for bcr-ab/, but only one patient relapsed. There was a 91%
concordance between PCR tests of simultaneously obtained
BM andperipheral blood.These analyses show that the
PCR
assay of the bcr-ablfusion transcript6 t o 12 months postBMT is an independentpredictorof
subsequent relapse
which provides an opportunity for early therapeutic intervention.
0 1995 b y The American Society of Hematology.
B
Philadelphia chromosome (Ph). This technique can identify
the bcr-ubl transcript at a dilution of a single Ph-positive
cell in a background of I O 5 to 10' normal cells. Although
the PCR assay can detect the presence of the bcr-ab1 transcript with exquisite sensitivity, the association of the detection of bcr-ab1 post-BMT and relapse is uncertain."-*' Several studies have examined the association of PCR-positivity
post-BMT, but the two largest studies are contradictory in
conclusions. Roth et a12' studied 64 patients post-BMT and
found that while PCR status early after transplant (< 100
days) did not correlate with relapse, PCR-positivity at later
periods did predict relapse. Overall 37 of 64 patients had at
least one positive assay, and 13 of these 37 (35%) relapsed.
When multiple samples were analyzed for each patient, those
with 2 1 PCR-positive assay had a 40% probability of relapse, and those with all PCR-positive assays had 70% risk
of relapse. Patients with multiple negative assays did not
relapse. Although this study concludes that PCR-positivity
at 6 months post-BMT is associated with relapse, the study
by Miyamura et al," coincidentally also with 64 patients,
suggests that PCR-positivity is not so foreboding. In their
study patients were from 2 to 220 weeks from BMT. The
analysis divided patients into two groups. Thirty-one patients
had at least one negative PCR test, and none of these patients
relapsed. In contrast, 33 patients had all PCR tests positive,
yetonly 7 relapsed. Moreover, 21 patients had a positive
PCR assay greater than 1 yearfrom transplant, withonly
two relapses.
Thus far, all studies have been too small to evaluate the
PCR assay results with other variables known to influence
the risk of relapse, such as the phase of disease, the type of
allograft, and the presence of GVHD. Does the PCR assay
have predictive power even when controlling for these factors, or does it merely recapitulate a known relapse risk? We
report here onthe PCR evaluation of 346 CML patients postBMT and the estimate of risk of relapse associated with
PCR-positivity for bcr-abl, controlling for other clinical vari-
ONE MARROW transplantation (BMT) is the only curative modality for chronic myeloid leukemia (CML),
but relapse occurs in 10% to 20%of patients receiving transplants in chronic phase"3 and over 50% of patients receiving
transplants in accelerated or blast c r i ~ i s . ' . ~
In. ~addition to
phase of disease, other clinical features that may influence
the risk of relapse include the type of donor allograft (unmanipulated BM v T-cell-depleted marrow), and the presence
or absence of acute or chronic graft-versus-host disease
(GVHD), with patients receiving unmanipulated marrow or
developing GVHD having a lower risk of relapse than those
receiving T-cell-depleted marrow or without GVHD.'."'
The ability to identify patients at high risk for relapse postBMT is of considerable interest because it might allow for
early interventions, including manipulations of the immunosuppression used to prevent and treat GVHD, the addition
of donor leukocytes,*"' or the addition of drugs such as ainterferon.'*
The polymerase chain reaction (PCR) can detect CML
cells through the amplification of the unique bcr-ab1 fusion
mRNA transcript, which is the molecular correlate of the
From the Divisions of Clinical Research, Experimental Pathology,
and the Program of Molecular Medicine, Fred Hutchinson Cancer
Research Center, Seattle, WA.
fDeceased.
Submitted August 12, 1994; accepted December 19, 1994.
Supported by Grants No. CA18029 and CA18221 from the National Cancer Institute.
Address reprint requests to Jerry Radich, MD, Fred Hutchinson
CancerResearchCenter,C2-023, 1124 Columbia St, Seattle, WA
98104.
The publication costsof this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1995 by The American Society of Hematology.
0006-497//95/8509-0012$3.00/0
2632
Blood, Vol 85, No 9 (May 1). 1995: pp 2632-2638
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
BCR-ABL IN CML POST-BMT
abies associated with relapse in multivariable models. We
find that a positive PCR assay at 6 to 12 months post-BMT
is an independent predictor of relapse after separately controlling for phase of disease, type of donor, and acute and
chronic GVHD. We also note that PCR positivity is less
common in patients greater than 12 months from transplant
and is less strongly associated with relapse. Like allogeneic,
related transplants, unrelated donor (URD) transplants are
commonly PCR-positive early after transplant but, in contrast, URD patients who are PCR-positive infrequently relapse.
MATERIALS AND METHODS
Patients. All patients receiving transplants for CML at the Fred
Hutchinson Cancer Research Center between 1976 and 1993 were
eligible for study. The transplant protocols were approved by the
Institutional Review Board. Patients were identified after BMT before their discharge home, or for patients already home, through
contact with our Long-Term Follow-Up (LTFW) program. Requests
were made to receive both BM and peripheral blood (PB) at 6month intervals. The LTFU staff were unaware of the results of
PCR assays. Many patients elected to have samples obtained at their
yearly LTFU check at our institution. These BM and PB samples
were used to perform cytogenetic evaluations and the PCR assay
for the presence of bcr-ab[ mRNA.
Transplant regimen. Patients receiving transplants in chronic
phase before 1990 received a conditioning regimen of cyclophosphamide (CY) 60 mgkg for 2 days (120 mgkg total) and total body
irradiation (TBI) at a dose of 200 cGy for 6 days (1,200 cGy total).’
Since 1990 chronic phase patients were randomized tothe above
regimen versus busulfan (BU) 16 mgkg and CY120
Accelerated- and blast-phase patients received a preparative regimen of
CY 120 m a g and TB1 of 225 cGy for 7 days (1,575 cGy total).’
Unrelated donor transplant recipients received a preparative regimen
of CY 120 m@g and either 1,200 cGy of T B 1 if in chronic phase
or 1,320 cGy if in accelerated or blast phase. All patients received
GVHD prophylaxis with methotrexate and cyclosporine as previously described.M
Sample preparation. RNA isolation from leukocytes from fresh
BM and PBsamples was performed by the acid guanidinium thiocyanate-phenol-chloroform method as previously described.” Aliquots
from this RNA were used for the detection of bcr-ab1 and the control
g 2 microglobulin mRNA (below).
Reverse transcription (RT) PCR assay for bcr-abl. We used a
two-step, “nested” RTPCR to amplify the chimeric bcr-ab1 mRNA.
The nested reaction yields either a 305-bp or a 234-bp PCR product
depending on the expression of bcr exon 3. One microgram of total
RNA was added to a combination 50 pL RTPCR mix of 25 pmol
of oligonucleotides CMLND and CMLNC, 50 mmoVL KCL, 10
mmom TRIS-HCL (pH 9.0). 0.1%Triton X-100 (Sigma, St Louis,
MO), 200 pmoVL dioxynucleotides (dNTPs), 1.5 mmol/L MgClz,
1.25 U Taq polymerase (Amplitaq; Perkin Elmer Cetus, Nonvalk,
CT), 10 U RNAsin, 7.5 U AMV RT (Boehringer Mannheim, Indianapolis, IN), and DEPC H20. This reaction mixture was incubated
for 30 minutes at 42”C, brought to 95°C for 5 minutes, and then
amplified for 40 cycles of 94°C for 30 seconds, 55°C for 1 minute,
and 72°C for 1 minute followed by a final elongation step at 72’C
for 7 minutes, using an Omigene thermocycler (Hybaid, Ltd, Woodbridge, NJ). Five microliters of this first reaction was transferred to
a 45-pL second reaction mixture containing fresh PCR ingredients
of25pmol oligonucleotides CMLNA and CMLNB, 500 mmoVL
KCl, 10 moVL TRIS-HCL (pH 9.0), 0.1% Triton X-100, 1 mmol/
L dNTPs, 1.5 mmol/L MgC12, 1.5 U Taq polymerase, and DEPC
2633
HzO. The tubes were placed into a preheated (65°C) thermocycler,
and after an initial denaturation step of 95°C for 5 minutes, 40 cycles
of amplification at94°C for 30 seconds, 55°C for 1 minute, and
72°C for 1 minute followed by a final elongation step at 72°C for 7
minutes was performed. After amplification 15 pL of the final PCR
product was electrophoresed through ethidium-bromide-stained 2%
agarose gel and photographed.
For all PCR reactions a
dilution of K562 RNA into the Phnegative cell line HL60 RNA served as a positive control. Negative
controls included both a no-RNA PCR mix (“blank”) and HL60
RNA. A separate amplification of p2 microglobulin was used as a
control ofRNA integrity andwas performed as previously described.%
Precautions to eliminate PCR carryover contamination included
separate rooms for pre-PCR and amplification procedures, filtercontaining disposable pipette tips, and no-nucleic acid PCR reactions
as negative control in all PCR amplification reactions.’* Using these
precautions we had an approximately 2% incidence of false positive
assays, defined as a positive PCR result in the no-RNA or HL60
RNA reactions. In these instances all samples run in the amplification
“batch” (usually 10 samples) was repeated.
The nucleotide sequences of the PCR primers used in bcr-ab1
amplification were: CMLNA 5’-TGGAGCTGCAGATGCTGACCAACTCG-3’; CMLNB 5“ATCTCCACTGGCCACAAAATCATACA-3’; CMLNC 5“GAAGTGTTTCAGAAGClTCTCC-3’;
CMLND 5”TGATTATAGCCTAAGACCCGGA-3’.
Definition of positive and negative PCR assays. The PCR assays
were performed without knowledge of the patient’s cytogenetic results or hematologic remission status. A positive PCR test required
a correct-size bcr-ab1 PCR product as well as negative “blank” and
HL60 amplifications. A negative assay required the absence of a
bcr-ab1 PCR product as well as no amplification of the “blank”
and HL60-negative controls, successful amplification ofthe
diluted K562-positive control, and a successful p2 microglobulin
PCR amplification. If a positive or negative control did not amplify
as expected, the entire panel of patient samples done in that amplification “batch” was discarded and repeated.
Definitions of outcomes and statistical methods. In this study
we defined relapse by both hematologic and cytogenetic criteria.
Cytogenetic relapse was defined as 2 5 metaphases positive for the
Ph chromosome atany point in time or the presence ofanyPh
chromosomes on two successive cytogenetic evaluations at least 6
months apart. Patients were classified as being PCR-positive or
-negative at the time interval of sampling. Relapse and survival rates
for patients appropriately classified at the stated time intervals were
estimated by the method of Kaplan and Meier.33For the endpoint
of relapse patients who died without relapse were censored at the
time of death. The hazard rates of relapse and survival for patients
classified as PCR-positive and PCR-negative at specific time intervals were compared using the log-rank test.’4 Multivariable analyses
were performed by fitting Cox regression models and considering
the data to be left truncated. This allowed the variable entry times
onto study to be ac~ommodated.’~
PCR status was considered as a
time-dependent covariate in these models.
If a patient had PCR results that were positive concurrently with
a sample which showed hematologic or cytogenetic relapse, then
such a patient was classified according to the PCR status at the last
test before relapse. For example, a patient with a negative PCR assay
at 6 months who at 12 months was FCR-positive for bcr-ab1 with
cytogenetics positive for the Ph chromosome is regarded in this
analysis as a patient who relapsed with a negative PCR assay at 6
months.
Patients were excluded from statistical analyses if (1) they had
evidence of hematologic or cytogenetic relapse simultaineously to,
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
RADICHET AL
2634
Table 1. Patient Characteristics
N = 346
Sex: 148 F1198 M
Age: Median 36 (range 2-60)
BMTtype:
Allogeneic related
Unrelated
Twin
Phase:
Chronic
Accelerated
Blast
2nd chronic
501
.-
p<0.0001
A5
240
96
10
262
56
16
12
(69%)
(28%)
(3%)
(76%)
115%)
(6%)
(3%)
40
35
g 30
0)
L
25
22 20
15
10
or before. their first PCR assay: or (2) they received a second transplant before their first PCR assay.
5
0
6-12 m
3m
RESULTS
Patient characteristics. The clinical characteristics of
the 346 patients included in these analyses are shown in
Table 1. Six hundred thirty-four samples of PB and/or BM
were successfully analyzed by PCR; in 15 other samples
RNA did not amplify from the PB and BM. One hundred
sixty-two patients had one PCR test performed, and 184 had
two or more tests performed. Eighty-two patients had their
first PCR sample at 3 months post-BMT, 146 patients at 6
to 12 months, 83 patients at 12 months to 3 years, and 35
patients at greater than 3 years post-BMT. The total number
of patients tested in each interval after BMT, categorized by
type of donor and phase of disease, is shown in Table 2.
The median follow-up for all patients from the time of first
PCR assay to either relapse or last contact is 16 months
(range, 1 to 47 months).
PCR-pnsitivih~and the risk qf srrbsequent relapse. The
percentage of patients testing positive for hcr-ah1 by the
PCR assay at 3 months, 6 to I2 months, > 12 months to 36
months, and >36 months, and the Kaplan-Meier estimates
of relapse for those testing PCR-positive versus PCR-negative are shown in Fig 1. Thirty-two patients were subsequently excluded from these analyses because they relapsed
>36 m
~12-36
m
Interval from BMT
Fig 1. The percentage of patients testing PCR-positive and the
Kaplan-Meier estimate of relapse associated with a PCR-positiveand
PCR-negativetest, defined by time from
BMT. The x axis represents
the time interval from BMT, whereas the y axis is percentage. ( W
The percentage of patients testing
PCR-positiveat each time interval.
(0)
The Kaplan-Meier estimate of relapse for PCR-positive patients
tested in the designated time interval from BMT. (R) The estimate
of relapse associated with a negative PCR-test in the time interval.
The P values compare the risk of relapse for patients PCR-positive
before or at the time of their first PCR test. Forty-five percent
(35/77) of patients tested at 3 months post-BMT were PCRpositive, and their Kaplan-Meier estimate of relapse at 3.5
years is 32%. whereas the Kaplan-Meier estimate of relapse
of those patients PCR-negative at 3 months is 24% ( P =
.27). The percentage of patients PCR-positive at 6 to 12
months was 26%. The Kaplan-Meier relapse estimates for
patients testing positive for hcr-ah1 at 6 to 12 monthsis
shown in Fig 2. Forthe 38 patients testing PCR-positive,
the Kaplan-Meier estimate of relapse at 4.5 years is 42%,
compared with 3% for the 112 PCR-negative patients (P <
Table 2. Patients TestingPCR-Positive and Time FromTransplant
0.5.
6-12
Variable rnos
Related
URD
Twin
Chronic
Accelerated
Blast
2nd chronic
301151
Total 381150
3 rnos
23/49
12/28
010
22/51
5115
516
315
35177
221103
14/42
215
281124
9/20
115
011
>12. <36
rnos
251110
4/39
112
251121
4/22
113
0/5
>36
rnos
o)
14/45
v)
017
-
016
0.4-
a
0
0,
E
0.3.
0
12/49
216
011
011
14/58
The number of patients testing PCR-positive (numerator) and the
total number of patients tested (denominator) in the s 3 months, >3
to s 1 2 months, > l 2 to s 3 6 months, and >36 months post-BMT is
shown in each cell of the table. The first three rows categorizes the
patients by the type ofdonor; the next four rowscategorizes patients
by the phase of disease. Patients who had tests in more than one
interval are included in the counts of each interval.
0
1
2
3
4
5
Years ofter BMT
Fig 2. The Kaplan-Meier estimates of relapse for patients testing
PCR-positive (N = 38)at 6 t o 12 months compared with patients who
were PCR-negative(N = 112) for bcr-ablmRNA. Tick marks represent
patients alive and still atrisk of relapse.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
2635
BCR-ABL IN CML POST-BMT
n
5617
0/20
5546
0120
5177
0/31
5152
0135
5026
4872
n
OD0
*
I
m0
*
W
0
5045
0120
n
W20
W20
0120
o
m
0120
0
0
8 .
8
OD0 0/120112
0/20
0120
*
019
8
0
OR0
0120
0
0120
0
0
-
o
m
0/20
0120
8
W
0120
4698
0
0120
8
8
I
0120
4862
0/20
0
m
0120
m
0
Fig 3. Follow-up of nine patients who had their first PCR-positive
assay greater than 36 months post-BMT. IO) A PCR-negative test;
( 8 )a PCR-positive assay.Below the PCR result is a numericalfraction
ofPhchromosome positive metaphasesltotal metaphases from a
concurrent cytogenetic exam. ”-“ Indicates that a cytogenetic sample was not received. “I“ indicatesthat thesample yielded an inadequate cytogenetic preparation for evaluation.
.mol).The median time from PCR-positivity to relapse was
7 months (range, 1.5 to 12 months). A separate Kaplan-
Meier estimation was performed for just the allogeneic-related transplants. Ninety-eight patients were tested, and 20
were PCR-positive. The Kaplan-Meier estimate of relapse
associated with a PCR-positive test at 6 to 12 months postBMT was 62%, compared with 3% for PCR-negative patients (P < .mol).
Thirty of 151 patients (20%)tested from 12 to 36 months
post-BMT were PCR-positive. Five of the PCR-positive and
one of the PCR-negative patients have relapsed (KaplanMeier estimates of relapse 25% v l%, P < .OOOl). The
median time from PCR-positivity to relapse was 7 months
(range, 0.5 to 20 months). The median follow-up of the
patients who tested PCR-positive but have not relapsed by
June 1, 1994 is 28 months from the first PCR-positive assay.
Fifteen of 58 patients (25%)tested greater than 36months
post-BMT tested PCR-positive, and one relapsed. No PCRnegative patient relapsed. The median follow-up of these
patients from their first PCR-positive test to present time is
22 months. Nine patients had their first positive PCR assay
detected greater than 36 months post-BMT, and none have
relapsed (Fig 3). The latest finding of a new positive PCR
assay following a previous negative PCR was at 48 months
(unique patient number [UPN] 5026, Fig 3).
We also examined the effect of serial PCR testing among
184 patients with 2 2 PCR tests.For patients with all positive
PCR tests, the Kaplan-Meier estimate of relapse is 78%.
For patients with all negative PCR tests, the Kaplan-Meier
estimate of relapse is 3%. For patients with both PCR-positive and PCR-negative tests, the relapse estimate is 22% (P
< .0001).
PCR-positivity in URDpatients. Because of the observation that URD BMT recipients have a lower relapse rate than
allogeneic-related transplant recipientsz6.*’ we performed a
separate statistical analysis on the 96 patients receiving an
URD BMT. Twelve of 28 (43%) URD patients were PCRpositive at 3 months, and 4 relapsed, compared with one
relapse in the 16 PCR-negative patients ( P = .lo, Fisher’s
exact test). Fourteen of 42 (33%) URD patients tested 6 to
12 months post-BMT were PCR-positive, and 2 relapsed,
compared with no relapses in the 28 PCR-negative patients
(Kaplan-Meier estimate of relapse 18% v 0%, respectively,
P = .11).
Multivariableanalyses of MRDandrelapse.
Clinical
features potentially associated with relapse were added individually to PCR status at 6 or 12 months in multivariable
analyses testing the association between relapse and PCR
status along with various individual clinical variables (Table
3). These clinical variables included the HLA matching status of the donor (twin or matched v URD or mismatched),
phase of disease at BMT (chronic phase v accelerated or
blast phase), the presence or absence of acute (grade 2 11)
and chronic GVHD after BMT, and the use of a TBI-containing preparative regimen. Because of the relatively small
number of relapses, it was feasible to include only two
covariates in any given model (PCR status plus one clinical
variable). PCR-positivity alone was associated with a 26fold risk of relapse. Match status and acute GVHD were
independently significantly associated with relapse. Patients
who received an allogeneic matched transplant had a relative
risk (m)
of relapse of 5.1 compared withURD or mismatched transplant recipients, and patients with acute GVHD
grades 0 or I had anRRof
relapse of 0.4 compared to
patients with grades 2 11. However, these variables did not
influence the strength of the risk associated with a PCRpositive assay in either multivariable model. The phase of
disease and the chronic GVHD status did not contribute to
the risk of relapse independent of a PCR-positive assay.
Table 3. Multivariate Analysis of Risk of Relapse for Patients
Tested at 6 or 12 Months Post-BMT
Variable
PValue
PCR-positive 26.1
PCR-positive
match’
and
5.1
PCR-positive
phaset
and
PCR-positive
and
PCR-positive
and
PCR-positive
and
cGVHDS
0.6
aGVHD§
TB111
0.6
<.0001
29.3
<.0001
,003
<,000
24.81
.43
<.0001
24.9
.30
<.0001
30.9
0.4
.o 1
<.0001
.20
RR
0.7
26.1
95% Cl
8.9-76.1
10.0-86.4
1.8-15.1
8.4-73.0
0.3- 1.7
8.5-73.0
0.2-1.9
10.4-91.5
0.2-0.8
9.0-76.2
0.3-1.3
Relative risks (RR) of relapse were estimated by placing the PCR
status (positive v negative) by itself into a multivariable model and
then adding an individual covariate
to the model. Thus an RR > 1
reflects a elevated risk of relapse associatedwith the presence of the
covariates defined as: (*), allogeneic matched related transplant v
mismatched orURD; (t),chronic phase vaccelerated,blast, or second
chronic phase; (*), presence v absence of chronic GVHD; (§l, acute
GVHD grades 2 II vgrades 0-1; and (11). use of a TBI-containing preparative regimen v chemotherapy only.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
2636
RADICH ET AL
The same clinical variables along with the PCR status
entered into a multivariable model in a step-up fashion, using
all patients studied over all time intervals after transplantation. The RRof PCR-positivity was 30 ( P < .000l), and
the only other variable entering the model was the HLA
match (RR = 5.6, P = .016).
A separate analysis was performed on the 216 patients
who received an allogeneic-related transplant in chronic
phase. A step-up regression analysis was also performed in
this group, with the variables of PCRand GVHD status.
PCR status was the only variable that entered the model,
with anRRof18
( P < ,0001). Neither acute nor chronic
GVHD entered the model.
The correlation of blood and BM samples. PB and BM
samples obtained on the same day post-BMT were evaluated
on S 14 samples. Concordance between the PB and BM PCR
result occurred in 468 samples (91%). Discordance with BM
PCR-positivity and peripheral blood PCR-negativity occurred in 6.5% of tests, whereas a negative BM PCR assay
with a positive PB assay occurred in 2.5% of tests.
DISCUSSION
Although the presence of the bcr-ab1 chimeric mRNA
has been studied in CML patients previously, the predictive
il. .')actof a positive PCR, especially in light of clinical variat es associated with relapse, has been unclear. Does the
PCR detection of bcr-ab1predict relapse even when controlling for factors such as GVHD and phase of disease? The
present study of 346 patients allowed usto evaluate the
predictive power of the PCR in light of other clinical variables as well as examine the significance of the PCR assay
at different time points after BMT. We found that a single
PCR assay positive for the bcr-ab1 chimeric mRNA was
associated with a high risk of subsequent relapse. The PCR
assay discriminated best at 6 to 12 months post-BMT, when
patients who were PCR-positive experienced a 42% risk of
relapse, compared with only 3% for those PCR-negative.
Moreover, the PCR assay was the strongest predictor of
relapse and remained highly associated with relapse in multivariable models when adjusted for other clinical variables
associated with relapse in CML. In the multivariable model
of all patients PCR-positivity carried an RR = 30, whereas
in the subgroup of those receiving an allogeneic, matched
transplant while in chronic phase, the RR = 18. In patients
sampled greater than 36 months after BMT,andinURD
transplants, a PCR-positive test was not strongly associated
with relapse. It should be emphasized that these data do not
meanthat clinical variables such as phase of disease are
unimportant in describing relapse risk, but instead indicate
that PCR-positivity is in itselfa strong independent predictor
of relapse.
Our finding of a 42% Kaplan-Meier estimate of relapse
in patients with PCR-positivity for bcr-ab1 at 6 to 12 months
post-BMT is in agreement with the study by Roth et al.23
Moreover, we found a similar influence of multiple PCR
testing on predicting relapse, with a 78% estimate of relapse
for patients with two or more positive PCR tests, a 22% risk
for patients with only one positive test, and a 3% risk for
patients with repetitive PCR-negative assays. However, we
have chosen to emphasize the predictive power of a single
test, thus avoiding potential biases associated with multiple
testing, such as controlling for patients who either did not
receive subsequent testing for a variety of reasons or who
received testing at a different interval.
We found that IS of S8 patients who were studied greater
than 36 months post-BMT had at least one PCR-positive
test. Nine of these 14 had their first PCRtest become positive
at greater than 36 months after at least one negative PCR
test (Fig 3). However, with a median follow-up since first
PCR-positivity of nearly 2 years none of these long-term
patients has relapsed. These data suggest that PCR-positivity
in these long-term patients is associated with a lower risk
of relapse, or predicts a considerably slower tempo of relapse, than PCR-positivity at 6 to 12 months post-BMT. This
interesting trend is most clearly demonstrated in Fig 1, which
shows that the relapse risk associated with a PCR-positive
assay steadily decreases from 42% at 6 to 12 months to less
than 10% for those patients testing positive at greater than
36 months after BMT. These data suggest that CML may
be a heterogeneous disease which may be more or less aggressive. Early PCR-positivity (6 or 12 months post-BMT)
may delineate a group of patients with more aggressive disease in whichthetempoof
PCR-positivity to relapse is
relatively quick. In contrast, patients who become PCR-positive late after BMT may have more indolent disease, which
may relapse at a slower pace after PCR-positivity, if at all.
Thus, although patients PCR-positive at 6 or 12 months the
mGdianof first PCR-positive test to relapse was 7 months,
for patients PCR positive at greater than 36 months postBMT, the median follow-up from time of first PCR-positive
test to last contact is 22 months. Previously the most extensive investigation of long-term CML patients was reported
by Guerrasio et aI,** who tested 27 patients at greater than
36 months post-BMT. Five of those 27 patients had at least
one PCR-positive test. One had no subsequent test, one had
a subsequent PCR-positive assay, and the remaining three
reverted to PCR-negativity. None of the five patients relapsed. These data combined with our experience suggest
that PCR-positivity in long-term patients may not carry so
high a risk of relapse compared with those patients testing
positive sooner after BMT. Further follow-up willbeimportant to see if these long-term PCR-positive patients are
truly unlikely to relapse, or merely relapse at a slower rate
than those patients with early PCR-positive tests.
We were interested in separately evaluating patients receiving an URD transplant because these patients experience relapse rates less than
half of that of allogeneic, related transplants
for CML.36.37 Interestingly, these patients have
a similar rate of
PCR-positivity to the aggregate CML group at 3 months and
6 to 12 months. The difference in the URD experience is that
the Kaplan-Meier estimateof relapse associated witha positive
PCR assay is relatively low, only 18%. There are few reports
with which to c 'mpare this experience. Cross et ai" report on
15 URD recipients, IO of whom were PCR-positive at some
time during the post-BMT course. They report only one cytogenetic relapse in this group. The low progression to relapse in
PCR-positive patients parallels the low risk of relapse experienced in all URD transplants for CML, and may be related to
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
2637
BCR-ABL IN CML POST-B'MT
the graft-versus-leukemia effect.&' Thus, the common occurrence of PCR-positivity early after transplantation implies that
the preparative regimen frequently does not completely eliminate the leukemic cells and that the subsequent lower relapse
rates in the URD recipients revolves around the itnmunomodulatory effect associated with GVHD.
Although our findings contrastwithseveralstudies
which suggest that PCR-positivity after BMT is not associated with a dramatic increase of relapse risk,14.1','9.25.29
most studies, including our own, agree
that persistently
negative PCR assays from 6 to 12 months after BMT place
the patient in a low-risk category. In our study, a negative
PCR test at 6 to 12 months post-BMT was associated with
only a 3% Kaplan-Meier estimate of relapse. At longer
intervals from transplant, the predictive power of a negative exam was even more dramatic, as no patient with a
negative test after 12 months has relapsed, with a median
follow-up of 19 months. Taken together, all studies agree
that a negative PCR test is associated with a very low risk
of relapse. However, the implications of an early positive
test arelessclear.This
may becausedlargely
by the
sundry differences between studies both in variables related to therapy (eg,thecomposition
of thecohortin
regard to phases of disease, unmanipulated v T-cell-depletedmarrow,preparativeregimens,andimmunosuppression regimen), and to PCR testing (eg, PCR test assay
conditions and sensitivity, frequency
and time points of
testing, and rigor of cytogenetic testing).
Our study has practical implications for the study and
treatment of CML patients post-BMT. First, there was a
median of approximately 6 months from PCR-positivity to
subsequent relapse, long enough to launch intervention trials.
Therapeutic approaches include the reduction in the amount
of immunosuppression given to prevent GVHD, the use of
a interferon," or the administration of donor buffy coat infusions."" Secondly, the high concordance between results
obtained from PB and BM shows that PB can be used for
monitoring. Thirdly, the PCR test has less predictive value
in long-term patients and URD recipients who may not benefit from routine testing for bcr-abl.
Recently studies using a semi-quantitative or competitive
PCR assay for the bcr-abl fusion mRNA24.26
have suggested
that a progressive increase in bcr-ab2 copy number can be
used to identify those patients who will relapse from those
who will not. Validation of these data may not improve on
the predictive power of a negative PCR test but could improve the predictive power of a positive test. However, caution may be required to interpret quantitative results from
center to center for reasons involving both center-specific
treatment and patient population as well as methods of sampling and PCR testing.
In summary, the PCR assay appears to discriminate between patients at a very low risk of relapse and those at a
higher risk of relapse. The predictive power of the PCR
assay appears independent of other relapse risk factors. Our
data indicate that the PCR test is of greatest predictive value
when performed at 6 to 12 months after BMT in allogeneicrelated donor recipients. PCR testing makes possible early
intervention to avert relapse.
REFERENCES
1. Thomas ED, Clift A, Fefer A, Appelbaum FR, Beatty P, Bensinger WI, Buckner D, Cheever MA, Deeg HG, Doney K,Flournoy
N, Greenberg P, Hansen J, Martin P, McGuffin R, Ramberg R,
Sanders JE, Singer J, Stewart P, Storb R, Sullivan K, Weiden PL,
Witherspoon R: Marrow transplantation for the treatment of chronic
myelogenous leukemia. Ann Intern Med 104:155, 1986
2. Goldman JM, Gale RP, Horowitz MM, Biggs JC, Champlin
RE, Gluckman E, Hoffman RG, Jacobsen SJ, Marmont AM,
McGlave PB, Messner HA, Rimm AA, Rozman C, Speck B, Tura
S, Weiner RN, Bortin MM: Bone marrow transplantation for chronic
myloid leukemia in chronic phase: Increased rate of relapse associated with T-cell depletion. AM Intern Med 108:806, 1988
3. Marks DI, Hughes TP, Szydlo R, Kelly S, Cullis JO,Schwarer
AP, Mackinnon S, Apperley J, Barrett J, Hows JM, Goldman JM:
HLA-identical sibling donor marrow transplantation for chronic myeloid leukaemia in first chronic phase: Influence of GVHD prophylaxis on outcome. Br J Haematol 81:383, 1992
4. Martin PJ, Clift RA, Fisher LD, Buckner CD, Hansen J, Appelbaum FR, DoneyKC, Sullivan KM, Witherspoon RP, Storb R,
Thomas ED: HLA-identical marrow transplantation during accelerated-phase chronic myelogenous leukemia: Analysis of survival and
remission duration. Blood 72:1978, 1988
5 . Clift RA, Buckner CD, Thomas ED, Bensinger WI, Bowden
R, Bryant E, Deeg HJ, Doney KC, Fisher LD, Hansen JA, Martin
P, McDonald GB, Sander JE, Schoch G, Singer J, Storb R, Sullivan
KM, Witherspoon RP, Appelbaum FR: Marrow transplantation for
chronic myeloid leukemia. A randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide. Blood 84:2036, 1994
6. Sullivan KM, Weiden PL, Storb R, Witherspoon RP, Fefer A,
Fisher L, Buckner CD, Anasetti C, Appelbaum FR, Badger C, Beatty
P, Bensinger W, Berenson R, Bigelow C, Cheever MA, Clift R,
Deeg HJ, Doney K, Greenberg P, Hansen JA, Hill R, Loughran T,
Martin P, Neiman P, Petersen FB, Sanders J, Singer J, Stewart P,
Thomas ED: Influence of acute and chronic graft-versus-host disease
on relapse and survival after bonemarrow transplantation from
HLA-identical siblings as treatment of acute and chronic leukemia.
Blood 73:1720, 1989
7. Horowitz MM, Gale RP, Sonde1 PM, Goldman JM, Kersey J,
Kolb H-J, Rimm AA, Ringden 0, Rozman C, Speck B, Truitt RL,
Zwaan FE, Bortin MM: Graft-versus-leukemia reactions after bone
marrow transplantation. Blood 75:555, 1990
8. Kolb HJ, Mittermuller J, Clemm Ch, Holler E, Ledderose G ,
Brehm G, Heim M, Wilmanns W: Donor leukocyte transfusions for
treatment of recurrent chronic myelogenous leukemia in marrow
transplant patients. Blood 76:2464, 1990
9. Cullis JO, Jiang YZ, Schwarer AP, Hughes TP, Barrett AJ,
Goldman JM: Donor leukocyte infusions for chronic myeloid leukemia in relapse after allogeneic bone marrow transplantation. Blood
79:1379, 1992
IO. Drogyski WR, Keever CA, Roth MS, Koethe S, Hanson G,
McFadden P, Gotttschall JL, Ash RC, van Tuinen P, Horowitz,
Flomenberg N: Salvage immunotherapy using donor leukocyte infusions as treatment for relapsed chronic myelogenous leukemia after
allogeneic bone marrow transplantation: Efficacy and toxicity of a
defined T-cell dose. Blood 82:2310, 1993
1 l . Szer J, Grigg AP, Phillips CL, Sheridan WP: Donor leukocyte
infusions after chemotherapy for patients relapsing with acute leukaemia following allogeneic BMT. Bone Marrow Transplant 11:109,
1993
12. Higano CS, Raskind WH, Singer JW: Use of alpha interferon
for the treatment of relapse of chronic myelogenous leukemia in
chronic phase after allogeneic bone marrow transplant. Blood
80:1437, 1992
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
2638
13. Morgan GJ, Janssen JWG, Guo AP, Wiedemann LM, Hughes
T, Gow J, Goldman JM, Bartram C R Polymerase chain reaction
for detection of residual leukaemia. Lancet 1:928, 1989
14. Gabert 3, Lafage M, Maraninichi D, Thuret I, Carcassonne
Y, Mannoni P Detection of residual bcr/abl translocation by polymerase chain reaction in chronic myeloid leukaemia patients after
bone-marrow transplantation. Lancet 2.1 125, 1989
15. Martiat P, Maisin D, Philippe MK, Ferrant A, Michaux JL,
Cassiman JJ, Van den Berghe H: Detection of residual BCWABL
transcripts in chronic myeloid leukaemia patients in complete remission using the polymerase chain reaction and nested primers. Br .l
Haematol 75:355, 1990
16. Kohler S, Galili N, Sclar JL, Donlon TA, Blume KG, Cleary
ML: Expression of bcr-ab1 fusion transcripts following bone marrow
transplantation for Philadelphia chromosome-positive leukemia.
Leukemia 4 5 4 1, 1990
17. Sawyers CL, Timson L, Kawasaki ES, Clark SS, Witte ON,
Champlin R: Molecular relapse in chronic myelogenous leukemia
patients after bone marrow transplantation detected by the polymerase chain reaction. Proc Natl Acad Sci USA 87563, 1990
18. Delfau MH, Kerckaert JP, d’Hooghe MC, Fenausx P, Lai JL,
Jouet JP, Grandchamp B: Detection of minimal residual disease in
chronic myeloid leukemia patients after bone marrow transplantation
by polymerase chain reaction. Leukemia 4: 1, 1990
19. Snyder DS, Rossi JJ, Wang JL, Sniecinski U, Slovak ML,
Wallace RB, Forman SJ: Persistence ofbcr-ab1 gene expression
following bone marrow transplantation for chronic myelogenous leukemia in chronic phase. Transplantation 51:103, 1991
20. Delage R, Soiffer RJ, Dear K, Ritz J: Clinical significance of
BCR-ABL gene rearrangement detected by polymerase chain reaction after allogeneic bone marrow transplantation in chronic myelogenous leukemia. Blood 78:2759, 1991
21. Hughes TP, Morgan GJ, Martiat P, Goldman JM: Detection
of residual leukemia after bone marrow transplantation for chronic
myeloid leukemia: Role of polymerase chain reaction in predicting
relapse. Blood 77:874, 1991
22. Guerrasio A, Martinelli G, Saglio G, Rosso C, Zacaria A,
Rosti G, Testoni N, Ambrosetti A, Izzi T, Sessargeo M, Frassoni F,
Gasparini P, Chiamenti A, Di Bartolomeo P, Pignatti PF: Minimal
residual disease status in transplanted chronic myelogenous leukaemia patients; low incidence of polymerase chain reaction positive
cases among 48 long disease free subjects who received unmanipulated allogeneic bone marrow transplantation. Leukemia 6507, 1992
23. Roth MS, Antin JH, Ash R, Valeri TH, Gotlieb M, Silver SM,
Ginsburg D: Prognostic significance of Philadelphia chromosomepositive cells detected by the polymerase chain reaction after allogeneic bone marrow transplant for chronic myeloid leukemia. Blood
791276, 1992
24. Thompson JD, Brodsky I, Yunis JJ: Molecular quantification
of residual disease in chronic myelogenous leukemia after bone
marrow transplantation. Blood 79: 1629, 1992
25. Lee M, Khouri I, Champlin R, Kanarjian H, Talpaz M, TrujilloJ, Freireich E, Deisseroth A, Stass S: Detection of minimal
residual disease by polymerase chain reaction of bcr/abl transcripts in
RADICH ET AL
chronic myelogenous leukaemia following allogeneic bone marrow
transplantation. Br J Haematol 82:708, 1992
26. Cross NCP, Feng L, Chase A, Bungey J, Hughes TP, Goldman JM: Competitive polymerase chain reaction to estimate the
number of bcr-ab1 transcripts in chronic myeloid leukemia patients
after bone marrow transplant. Blood 82:1929, 1993
27. Cross NCP, Hughes TP, Lin F, O’Shea P, Bungey .I, Marks
DI, Ferrant A, Martiat P, Goldman JM: Minimal residual disease
after allogeneic bone marrow transplantation for chronic myeloid
leukaemia in first chronic phase: Correlations with acute graft-versus-host disease and relapse. Br J Haematol 84:67. 1993
28. Arnold R, Janssen JWG, Heinze B, Bunjes D, Hertenstein B,
Wiesneth M,Kubanek B, HeimpelH, Bartram CR: Influence of
graft-versus-host disease on eradication of minimal residual leukemia detected by polymerase chain reaction in chronic myeloid leukemia patients after bone marrow transplantation. Leukemia 7:747,
1993
29. Miyamura K, Tahara T, Tanimoto M, Morishita Y, Kawashima K, Morishima Y, Saito H, Tsuzuki S, Takeyama K, Kodera
Y, Matsuyama K, Hirabayashi N, YamadaH, Naito K,Imai K,
Sakamaki H, Asai 0, Miqutani S: Long peristent bcr-ab1 positive
transcript detected by polymerase chain reaction after marrow transplant for chronic myelogeous leukemia without clinical relapse: A
study of 64 patients. Blood 81:1089, 1993
30. Storb R, Deeg HJ, Whitehead J, Appelbaum F, Beatty P,
Bensinger W, Buckner CD, Clift R, Doney K, Farewell V, Hansen
J, Hill R, Lum L, Martin P, McGuffinR, Sanders J, Stewart P,
Sullivan K, Witherspoon R, Yee G, Thomas ED: Methotrexate and
cyclosporine compared with cyclosporine alone for prophylaxis of
acute graft versus host disease after marrow transplantation for leukemia. N Engl J Med 313:729, 1986
31. GehlyGB, Bryant EM, Lee AM,KiddPG,
Thomas ED:
Chimeric BCR-ab1 messenger RNA as a marker for minimal residual
disease in patients transplanted for Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood 78:458, 1991
32. Kwok S, Higushi R: Avoiding false positives with PCR. Nature 339:237, 1989
33. Kaplan EI, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457, 1958
34. Kalbfleish JD, Prentice RI: The Statistical Analysis of Failure
Time Data. New York, NY, Wiley, 1980
35. Cox DR, Oakes D: Analysis of Survival Data. London, UK,
Chapman and Hall, 1984
36. McGlave P, Bartsch G, Anasetti C, Ash R, Beatty P, Gajewski
J, Kernan NA: Unrelated donor marrow transplantation therapy for
chronic myelogenous leukemia: Initial experience of the National
Marrow Donor Program. Blood 81543, 1993
37. Drobyski WR, Ash RC, Casper JT, McAuliffe T, Horowitz
MM, Lawton C, Keever C, Baxter-Lowe LA, Camitta B, Garbrecht
F, Pietyga D, Hansen R, Chitambar CR, Anderson T, Flomenberg
N: Effect of T-cell depletion as graft-versus-host disease prophylaxis
on engraftment, relapse, and disease-free survival in unrelated marrow transplantation for chronic myelogenous leukemia. Blood
83:1980, 1994
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
1995 85: 2632-2638
Polymerase chain reaction detection of the BCR-ABL fusion
transcript after allogeneic marrow transplantation for chronic myeloid
leukemia: results and implications in 346 patients
JP Radich, G Gehly, T Gooley, E Bryant, RA Clift, S Collins, S Edmands, J Kirk, A Lee and P
Kessler
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/9/2632.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.