Monitoring Minimal Residual Disease and Predicting Relapse in APL By

Leukemia (2001) 15, 1060–1065  2001 Nature Publishing Group All rights reserved 0887-6924/01 $15.00 www.nature.com/leu Monitoring minimal residual disease and predicting relapse in APL by quantitating PML-RARα transcripts with a sensitive competitive RT-PCR method K Tobal, H Moore, M Macheta and JA Liu Yin

Molecular Oncology Group, University Department of Haematology, Manchester Royal Inﬁrmary, Manchester M13 9WL, UK

Qualitative RT-PCR methods used for monitoring minimal for diagnosis and in monitoring minimal residual disease residual disease (MRD) in APL patients fail to predict relapse (MRD).5–12 Serial negative tests for the PML-RARα transcripts in up to 25% of patients in remission. We report here the devel- opment and evaluation of a highly sensitive (10−5 and 10−6 with in post-chemotherapy patients are generally associated with one round and two rounds of PCR, respectively) competitive prolonged remission. However, up to 25% of those patients RT-PCR method to quantitate the PML-RARα fusion transcripts. can relapse within a short period of testing negative.10–13 This PML-RARα transcript’s levels were normalised to 105 copies of is due in part to the low level of sensitivity of these methods, ABL transcript. Serial BM and PB samples from 16 patients with approximately 10−4, which is 2 log lower than corresponding APL and t(15;17) were examined. Presentation samples from methods for other fusion genes such as AML1-MTG8.14 These three patients (three BM, one PB) showed levels in the range of 0.7 × 106–3.5 × 106 and 1.2 × 105 molecules in BM and PB methods lack the necessary sensitivity to detect the presence samples respectively. Serial quantitation of MRD in both BM of low levels of leukaemic cells present during stable and PB samples showed significantly lower levels of PML- remission. RARα transcripts in remission, although the majority of It is clear from the data reported by various groups on MRD samples remain positive for the PML-RARα transcripts even monitoring of APL that a sensitive quantitation of PML-RARα those in long-term remission (up to 94 months). Levels of PML- may provide a superior strategy for MRD monitoring in these RARα in remission samples were up to 2 × 102 and up to 5.2 × 1 patients. Recently, protocols for real-time quantitation of PML- 10 molecules in BM and PB respectively. BM and PB samples α taken from two patients 2–4 months before relapse showed sig- RAR have been developed, but these small studies have pro- nificantly higher levels of PML-RARα transcripts (1.2 × 104 mol- duced mixed results.15,16 Evaluation of real-time quantitation ecules in BM; 3.5 × 102, 1.2 × 102 and 1.2 × 103 in PB). The same of PML-RARα by various groups is in progress. samples, when tested with a standard qualitative RT-PCR for We have developed a highly sensitive competitive RT-PCR α −4 the amplification of PML-RAR (with a sensitivity of 10 ) pro- method for the quantitation of PML-RARα transcripts and have duced negative results. This indicates that the qualitative methods would not have predicted relapse in these patients. assessed its value in monitoring MRD in 16 patients with APL Our data show that quantitating PML-RARα transcripts with a and t(15;17). sensitive method may provide a superior approach for monitoring MRD in APL and identifying patients at high risk of relapse. Leukemia (2001) 15, 1060–1065. Keywords: APL; t(15;17); PML-RARα; minimal residual disease; competitive RT-PCR Materials and methods

Introduction Patients Acute promyelocytic leukaemia (APL) is characterised by the α 1–3 t(15;17), which produces the PML-RAR fusion gene. Tran- Serial bone marrow (BM) and peripheral blood (PB) samples α scripts for PML-RAR are detected in all patients with t(15;17). were collected from 16 patients in complete remission (CR) The breakpoints on chromosome 17 were located in the of APL with a CR duration of 9–94 months (median 32 second intron of the RARA gene in all cases with t(15;17), months). Three of the patients were tested at presentation. whereas there are three different breakpoint cluster regions Diagnosis of APL (AML M3) was made according to FAB (BCR1, BCR2, BCR3) on chromosome 15. This leads to the classification17 and the presence of t(15;17) was confirmed by α formation of three types of PML-RAR transcripts. BCR1 and karyotypic analysis. All patients received standard intensive BCR3 are most frequently involved in APL (92–95% of cases) remission induction chemotherapy with an anthracycline/ α and produce a unique PML-RAR fusion gene, whereas BCR2 anthracenedione (daunorubicin or mitozantrone), cytosine leads to several different products due to different breakpoints arabinoside ± thioguanine and two courses of consolidation 4 within this cluster. The combination of all-trans retinoic acid chemotherapy with the same agents, followed by autologous (ATRA) and chemotherapy has made it possible for up to 90% or allogeneic bone marrow transplant (BMT) in four patients. of patients to achieve complete remission (CR). However, Patients who did not receive BMT received a fourth course of relapse still occurs in some 15–25% of patients, indicating the intensive chemotherapy with mitozantrone and an intermedi- need for an accurate protocol for the monitoring of minimal ate dose of cytosine arabinoside. Three patients received short residual disease (MRD) in these patients. Such a protocol may courses of ATRA 5–7 days prior to induction chemotherapy. enable the identification of patients at high risk of relapse and Five patients received concomitant ATRA with induction predict its onset at an early stage. RT-PCR methods developed chemotherapy until achievement of remission. Pre-BMT con- α for the amplification of PML-RAR transcripts have been used ditioning was comprised of daunorubicin (60 mg/m2), cyclo- phosphamide (120 mg/m2) and TBI (1000 cGy) in two frac- tions for allograft or melphalan (120 mg/m2) and TBI for Correspondence: K Tobal; Fax: 0161-2764088 autograft. The remission BM karyotypes were normal in all Received 26 October 2000; accepted 26 March 2001 patients. Monitoring MRD in APL with a competitive RT-PCR K Tobal et al 1061 Samples and RNA preparation fication parameters were 95°C for 12 min (one cycle); 93°C for 1 min, 56°C for 1 min and 72°C for 1 min (50 cycles); and Mononuclear cells (MNCs) from PB and BM samples were 72°C for 5 min (one cycle). Two µl of first round product were obtained by Ficoll–Hypaque density gradient centrifugation used in a 50 µl second round PCR reaction containing primers and either stored at −80°C or used immediately for RNA iso- NP32 and RAR for BCR1 and NP5 and RAR for BCR3. PCR lation. Between 1 × 106 and 2 × 106 cells from the NB4 cell parameters were 95°C for 12 min (one cycle); 93°C for 1 min, line or patients’ MNCs were used for RNA isolation by the 56°C for 1 min and 72°C for 1 min (40 cycles); and 72°C for guanidinium-phenol-chloroform method of Chomczynski and 5 min (one cycle). Second round products were electrophor- Sacchi18 with minor modifications as described previously.19 esed on a 2.5% agarose gel.

RT reaction ABL competitive RT-PCR

Four µl (1–5 µg) of RNA was heated at 72°C for 5 min and A competitive RT-PCR method was used to quantitate the ABL snap cooled on ice. RT reaction mixture (final concentration transcript.19 This quantitation was used to accurately estimate in a 20 µl reaction volume: 1× first strand buffer, 10 mM DTT, the level of amplifiable RNA present in samples. Two µlof 0.25 µg pd(N)6, 1 mM dNTPs, 200 U MMLV, 40 U RNAsin) cDNA were mixed with 2 µlofABL transcript competitor was then added to the RNA and incubated at room tempera- DNA and subjected to a one round PCR amplification as ture for 10 min. RT reaction was carried out at 37°C for 1 h, described above. The expected band sizes for ABL and 45°C for 30 min and 72°C for 5 min. competitor are 276 bp and 235 bp, respectively.

Control gene (ABL RT-PCR) RARα competitive RT-PCR

To assess the quality and estimate the quantity of amplifiable Competitors for BCR1 and BCR3 types of PML-RARα tran- RNA used for RT-PCR, qualitative and quantitative RT-PCR scripts were prepared by PCR amplification of the normal amplification of ABL gene transcripts was performed. PCR RARα transcripts with primers cpmr5 and 7 or cpmr3 and 7. amplification was performed as previously described;20 The sense primers, cpmr5 and cpmr3, have tails containing briefly, PCR amplification was performed in a 25 µl reaction sequences of primers NP3 and NP32 or NP51 and NP5 that containing primers CA3 and A2 at 97°C for 1 min 30 s, 64°C are used in PCR amplification respectively (Figure 1). None of for 50 s and 72°C for 1 min, (one cycle); 97°C for 30 s, 64°C our patients have a BCR2 fusion transcript, but a similar for 50 s and 72°C for 1 min (40 cycles) and 72°C for 5 min approach could be used to prepare BCR2 competitors. Three (one cycle). PCR products were electrophoresed on a 2% µlofPML-RARα positive samples’ cDNA were mixed with 2 agarose gel. µl of competitor DNA and subjected to PCR amplification as described above. Each sample was quantified at every order of magnitude and then at every half order of magnitude. Reac- PML-RARα RT-PCR tions that produce equal intensity bands for the transcripts and competitor were estimated to contain 1.7× the number of Three µl of cDNA were subjected to two rounds of transcript- competitor molecules in the reaction (the size of specific PCR amplification for the PML-RARα transcripts. The competitor/transcript). The level of PML-RARα present in first round PCR was performed in a 50 µl reaction containing 1 U of Amplitaq gold DNA polymerase (Perkin Elmer, War- rington, UK) and primers NP2 and 8 for BCR1 transcript, or primers NP51 and 8 for BCR3 transcript (Table 1). PCR ampli-

Table 1 Sequence of primers used in this study

Primer Sequence 59–39

A2 TTCAGCGGCCAGTAGCATCTGACTT CA3 TGTTGACTGGCGTGATGTAGTTGCTTGG NP2 AGTGGCGCCGGGGAGGCAGCCA NP32 GGCGCCGGGGAGGCAGCCA NP51 TCTTGCATCACCCAGGGGAAAGCC NP5 TGCATCACCCAGGGGAAAGCC CPMR3 GCCAGTGGCGCCGGGGAGGCAGCCACCAGCTTCC AGTTAGTGGA CPMR5 AGCTCTTGCATCACCCAGGGGAAAGCCACCAGCTT CCAGTTAGTGGA 7 TCTTCTGGATGCTGCGGCGG 8 GGCGCTGACCCCATAGTGGT RAR AGGGCTGGGCACTATCTCTTC M4 AGCTGCTGGAGGCCTGTGGACGCGCGGTACC 2 TGTGCTGCAGCGCATCCGCA R8 CAGAACTGCTGCTCTGGGTCTCAAT Figure 1 Schematic diagram of the synthesis of competitors for the BCR1 and BCR3 PML-RARα transcripts.

Leukemia Monitoring MRD in APL with a competitive RT-PCR K Tobal et al 1062 samples was normalised to the number of copies of PML- RARα transcripts/105 copies of ABL transcript.

ABL and PML-RARα degradation rates

Samples from two patients and the NB4 cell line were divided into three parts: one part was Ficolled as described above then frozen directly at −80°C; the other two parts were incubated at room temperature for 24 and 48 h, respectively, before being Ficolled. The levels of ABL and PML-RARα transcripts were estimated in all three portions to estimate the rate of degra- − ° dation. In one experiment all three parts were frozen at 80 C Figure 2 Sensitivity assay of the transcript-speciﬁc ampliﬁcation of before proceeding with the estimation of the levels of ABL and PML-RARα. N, negative control; M, molecular weight marker. PML-RARα transcripts.

and 10−6 after two rounds of PCR (Figure 2). This method can Reproducibility and accuracy of assays detect as few as three copies of the PML-RARα transcript.

In all tests, negative and positive controls were used. Negative controls included reaction with no RNA, no cDNA or t(15;17)- Linearity and reproducibility of results negative cell lines, such as K562. The NB4 cell line was used as a positive control. All necessary precautions were taken Serial dilutions of NB4 cell line were prepared with normal to avoid contamination. These included the use of especially MNCs and subjected to competitive RT-PCR analysis to assess designed UV-flow cabinet, PCR-designated pipettes and fil- the accuracy and reproducibility of results achieved with this tered tips for all PCR preparations. Tests were repeated twice method. Our data show this method to be accurate and linear to confirm the results. over a wide range of PML-RARα transcript levels. The points of equivalence for 10−6,10−5,10−4,10−3,10−2,10−1 of NB4 cells were 7, 7 × 101,7× 102,7× 103,7× 104,7× 105 copies Sensitivity analysis of competitor, respectively. Figure 3 shows an example of this analysis. All experiments were repeated to confirm the levels Normal MNCs were used to make serial dilutions from the of transcripts in each sample. NB4 cell line,21 which has a BCR1 PML-RARα fusion gene, or a presentation sample from an APL patient with a BCR3 PML-RARα. These dilutions were used in comparing sensitivity of different methods for the amplification of PML-RARα.

Standard qualitative RT-PCR ampliﬁcation of PML- RARα transcripts22

Three µl of cDNA were added to a final volume of 50 µlof first round PCR reaction buffer containing as final concen- trations: 1× PCR buffer (Gibco-BRL, Paisley, UK), 0.2% W-1

(detergent; Gibco-BRL), 1.5 mM MgCl2, 0.25 mM dNTPs, 15 pmol of primers M4 and RAR (Table 1), 2 U Taq DNA polymerase. Amplification was performed for one cycle at 94°C for 2 min, followed by 40 cycles at 93°C for 1 min, 55°C for 1 min, 72°C for 1 min, and finally one cycle at 72°C for 5 min. Five µl of first round PCR product was added to a final volume of 50 µl of second round reaction buffer containing 15 pmol of primers 2 and R8 (Table 1). PCR amplification was carried out as for the first round. Second-round PCR products were electrophoresed on a 2% agarose gel.

Results

Sensitivity analysis

Sensitivity of the transcript-speciﬁc ampliﬁcation of both BCR1 and BCR3 PML-RARα was 10−5 after one round of PCR Figure 3 Linearity studies of competitive PCR.

Leukemia Monitoring MRD in APL with a competitive RT-PCR K Tobal et al 1063

Figure 4 Serial quantitation of PML-RARα transcripts in BM and PB samples from seven patients with APL. Solid lines represent BM samples, while dotted lines represent PB samples. Arrows indicate samples taken up to 4 months before relapse. Numbers in the key represent individual patients UPN. Patients 1 and 2 experienced clinical relapse and are identiﬁed by the same numbers in Figure 6.

ABL and PML-RARα degradation rates

The degradation rates of ABL and PML-RARα transcripts were equal. Both transcripts decreased by 0.5 log after 24 h and 1 log after 48 h. RNA and cDNA samples stored at –80°C for up to 2 months showed no signiﬁcant degradation of either transcript.

PML-RARα quantitation

The level of PML-RARα transcripts was quantified in sequen- tial BM and PB samples from 16 patients with APL. Two of the 16 patients who achieved remission subsequently relapsed. We have examined three patients at presentation of their disease (three BM, one PB). The levels of PML-RARα transcripts at presentation were found to be in the range of 0.7 × 106–3.5 × 106 molecules in BM samples and 1.2 × 105 molecules in PB samples (Figures 4 and 5). Quantitation of MRD by RT-PCR analysis in 16 patients showed a significant reduction in the level of PML-RARα transcripts after the induction of CR as compared to levels at presentation in both BM (11 patients) and PB (eight patients). α α Figure 5 Levels of PML-RAR and PB samples from different While the level of PML-RAR transcripts became undetect- phases of APL in BM and PB samples from 16 patients. P, presentation able in two patients during durable remission, the majority of (Κ, λ); CR, remission (Π, σ); CR/BR, remission/before relapse (Γ, γ); patients continue to express detectable levels of fusion tran- R, relapse (ρ, ε). scripts even in long-term remission (9–94 months, median 32 months). There were no significant differences in the level of ABL transcripts/reaction between negative and positive samples. There was no difference in the levels of PML-RARα sample in second remission showed a significantly higher during CR between patients who received different treatment level of fusion transcript (1.2 × 104) than those detected during (chemotherapy or BMT). In all patients in stable remission the stable remission. This patient relapsed 4 months later (Figure level of PML-RARα was found to be up to 2 × 102 molecules 4, patient 2). Three PB samples taken between 2 and 3 months in BM samples and up to 5.2 × 101 molecules in PB samples before relapse also showed significantly higher levels of PML- (Figures 4 and 5). BM examination for patients in remission RARα transcripts (3.5 × 102, 1.2 × 102 and 1.2 × 103) than showed that samples were both morphologically and karyo- those detected in samples from patients in stable remission typically normal. (Figure 4, patients 1 and 2). When we examined the same Examination of the level of PML-RARα transcript in one BM samples, using the same cDNAs, with a standard qualitative

Leukemia Monitoring MRD in APL with a competitive RT-PCR K Tobal et al 1064 RT-PCR strategy for the quantitation of MRD in t(8;21) AML can identify patients in durable remission and predict clinical relapse.19 Protocols have been developed in recent years to quantitate PML-RARα transcripts using real-time quantitation stra- tegies.15,16 The value of these methods in MRD monitoring has not been fully assessed yet. We have described here a highly sensitive competitive RT-PCR method for the amplification and quantitation of PML-RARα transcripts. The use of transcript-specific RT-PCR for the various types of PML-RARα transcripts enabled us to avoid the amplification of unnecessary PML sequences. We have shown in previous work that the inclusion of these PML sequences reduces the Figure 6 Diagram showing a comparative study of MRD monitor- sensitivity of PML-RARα detection.22 Using this strategy, we ing with a standard qualitative RT-PCR protocol and the competitive were able to achieve a sensitivity of 10−5 with one round of RT-PCR in two patients with APL who relapsed. Standard qualitative −6 RT-PCR, (σ) PB negative; (Π) PB positive, (ε) BM negative; (ρ)BM PCR and 10 with two rounds of PCR. This sensitivity is com- positive. Competitive RT-PCR, (∀) significant increase in the level of parable to that achieved for other fusion transcripts such as PML-RARα detected by competitive RT-PCR. AML1-MTG8.14 Tests performed on serial dilutions of NB4 cell line showed this method to have equal efficiency of − amplification for the PML-RARα transcripts and the competi- RT-PCR protocol, with a sensitivity of 10 4 (see Materials and methods), they were found to be negative for PML-RARα tors. These tests also showed the quantitation to be linear over (Figure 6). a wide range of fusion transcripts level. The level of PML-RARα transcripts increased further at Our data showed the persistence of leukaemic cells α relapse. Two relapsed BM samples showed levels of 1 × 106 expressing the PML-RAR transcripts in long-term remission and 1 × 107 molecules of PML-RARα, while two relapsed PB patients and indicate that disease-free remission is not incon- samples showed levels of 1.2 × 105 and 3.5 × 105 molecules sistent with the presence of a small number of leukaemic cells. of PML-RARα (Figures 4 and 5). These data are in accordance with the findings we reported earlier.22 Quantitation of PML-RARα transcripts showed a significant Discussion decrease in the number of PML-RARα transcripts after chemotherapy, which imply a reduction in the number of leukaemic APL is highly responsive to ATRA and to chemotherapy, and cells, when patients respond to treatment. has a favourable clinical outcome. However, relapse remains Levels of fusion transcripts in presentation samples were in a major problem. Predicting early relapse in patients may pro- the range of 0.7 × 106–3.5 × 106 molecules in BM samples vide the opportunity to offer these patients additional or alter- and 1.2 × 105 molecules in PB samples. In stable remission native treatment such as ATRA, Aasenic trioxide or BMT. the levels of PML-RARα were significantly lower, but remain Monitoring MRD in APL has not been efficient in identifying detectable in the majority of samples, even though BM exam- all patients at high risk of relapse. To date, monitoring MRD ination for patients in remission showed that samples were in APL has been based on the detection of PML-RARα tran- both morphologically and karyotypically normal. Up to 2 × scripts with qualitative RT-PCR methods that have a low sensi- 102 molecules were detected in BM samples and up to 5.2 − tivity level of 10 4. This may explain the failure of these × 101 molecules were detected in PB samples during stable methods to detect early significant changes in the level of remission. Our data show that PML-RARα levels increased sig- α PML-RAR transcripts before the onset of haematological nificantly up to 4 months before the onset of clinical relapse relapse. This failure was most noticeable in patients who in two patients. Our examination of one BM sample, 4 months α relapse within short periods of negative PML-RAR tests with before relapse, showed 1.2 × 104 molecules of PML-RARα. these protocols. Up to 25% of patients tested negative with Three PB samples taken between 2 and 3 months before 10–13 these protocols relapsed shortly afterwards. Large studies relapse also showed significantly higher levels of PML-RARα that used qualitative RT-PCR protocols for MRD monitoring transcripts (3.5 × 102, 1.2 × 102 and 1.2 × 103 molecules). 23 have produced variable data. Diverio et al showed that the Examining the same samples with a standard qualitative RT- relapse rate was 95% in patients who are PCR positive during 24 PCR protocol (see Materials and methods section) showed remission and 5% in PCR-negative patients. Burnett et al, these samples to be negative for PML-RARα. This examination on the other hand, showed that the relapse rate was 57% in indicates that the qualitative protocol could not have PCR-positive patients and 27% in PCR-negative patients. predicted the clinical relapse experienced by these patients. These findings indicate that a sensitive RT-PCR method In summary, our results on this small series of patients with coupled with a quantitative approach for PML-RARα tran- APL and t(15;17) have shown that a sensitive and accurate scripts is required for efficient monitoring of MRD in APL. α Such quantitative protocols may provide a superior approach quantitation of PML-RAR fusion transcripts may provide a for identifying patients at high risk of relapse, at an early stage, superior approach for monitoring MRD in these patients than who then may be offered additional or alternative treatment the qualitative RT-PCR approach used to date. We suggest that to prevent the onset of haematological relapse. Indeed pre- PB samples may provide an alternative source for monitoring liminary data, reported by Lo Coco et al,25 indicate that MRD in APL patients, which may help in avoiding the incon- α pre-emptive administration of salvage therapy at the time of venience of BM sampling. The efficiency of PML-RAR quan- molecular relapse is beneficial in APL. titation in identifying patients at high risk of relapse requires Recently, we have shown that using a sensitive competitive further evaluation in larger series of patients.

Leukemia Monitoring MRD in APL with a competitive RT-PCR K Tobal et al 1065 Acknowledgements 12 Grimwade D, Howe K, Langabeer S, Burnett A, Goldstone A, Solo- mon E. Minimal residual disease detection in acute promyelocytic leukemia by reverse-transcriptase PCR: evaluation of PML-RARα This work was supported by the Manchester Leukas-Aid and RARα-PML assessment in patients who ultimately relapse. Research Charity. MM was an LRF Clinical Fellow. Leukemia 1996; 10: 61–66. 13 Devaraj PE, Foroni L, Prentice GH, Hoffbrand VA, Secker-Walker LM. Relapse of acute promyelocytic leukemia follows serial negative RT-PCR assays: a cautionary tale. Leuk Res 1996; 20: 733– References 737. 14 Tobal K, Liu Yin JA. Monitoring of minimal residual disease by 1 Rowley JD, Golomb HM, Dougherty C. 15/17 translocation, a quantitative reverse transcriptase-polymerase chain reaction for consistent chromosomal change in acute promyelocytic leu- AML1-MTG8 transcripts in AML-M2 with t(8;21). Blood 1996; 88: kaemia. Lancet 1977; 8010: 549–550. 3704–3709. 2 de The H, Chomienne C, Lanotte M, Degos L, Dejean A. The 15 Cassinat B, Zassadowski F, Balitrand N, Barbey C, Rain JD, Fenaux t(15;17) translocation of acute promyelocytic leukaemia fuses the P, Degos L, Vidaud M, Chomienne C. Quantitation of minimal retinoic acid receptor α gene to a novel transcribed locus. Nature residual disease in acute promyelocytic leukemia patients with 1990; 347: 558–561. t(15;17) translocation using real-time RT-PCR. Leukemia 2000; 14: 3 Kakizuka A, Miller WH Jr, Umesono K, Warrell RP Jr, Frankel SR, 324–328. Murty VVVS, Dmitrovsky E, Evans RM. Chromosomal translo- 16 Grimwade D, Diverio D, Harrison G, Wheatley K, Rodgers J, Lo cation t(15;17) in human acute promyelocytic leukemia fuses Coco F, Goldstone AH, Solomon E, Burnett AK. Detection of mini- RARα with a novel putative transcription factor, PML. Cell 1991; mal residual disease (MRD) in APL by ‘real-time’ RT-PCR: Analysis 66: 663–674. of cases entered into the UK MRC ATRA trial. Blood 1999; 94 4 Pandolfi PP, Alcalay M, Fagioli M, Zangrilli D, Mencarelli A, (Suppl. 1): 2778. Diverio D, Biondi A, Lo Coco F, Rambaldi A, Grignani F, Roch- 17 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DAG, ette-Egly C, Gaube MP, Chambon P, Pelicci PG. Genomic varia- Gralnick HR, Sultan C. Proposed revised criteria for the classi- bility and alternative splicing generate multiple PML-RARA tran- fication of the acute myeloid leukemia. Ann Intern Med 1985; scripts that encode aberrant PML proteins and PML-RARA isoforms 103: 620–625. in acute promyelocytic leukemia. EMBO J 1992; 11: 1397–1407. 18 Chomczynski P, Sacchi N. Single-step method of RNA isolation by 5 Borrow J, Goddard AD, Gibbons B, Katz F, Swirsky D, Fioretos T, acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Dube I, Winfield DA, Kingston J, Hagemeijer A, Rees JKH, Lister Biochem 1987; 162: 156–159. TA, Solomon E. Diagnosis of acute promyelocytic leukaemia by 19 Tobal K, Newton J, Macheta M, Chang J, Morgenstern G, Evans RT-PCR: detection of PML-RARA and RARA-PML fusion tran- PAS, Morgan G, Lucas GS, Liu Yin JA. Molecular quantitation of scripts. Br J Haematol 1992; 82: 529–540. minimal residual disease in acute myeloid leukaemia with t(8;21) 6 Tobal K, Saunders MJ, Grey MR, Liu Yin JA. Persistence of RARα- can identify patients in durable remission and predict clinical PML fusion mRNA detected by reverse transcriptase polymerase relapse. Blood 2000; 95: 815–819. chain reaction in patients in long-term remission of acute promye- 20 Cross NCP, Hughes TP, Feng L, O’Shea P, Bungey J, Marks DI, locytic leukaemia. Br J Haematol 1995; 90: 615–618. Ferrant A, Martiat P, Goldman JM. Minimal residual disease after 7 LoCoco F, Diverio D, Pandolfi PP, Biondi A, Rossi V, Avvisati G, allogeneic bone marrow transplantation for chronic myeloid leu- Rambaldi A, Arcese W, Patti MC, Meloni G, Mandelli F, Grignani kaemia in first chronic phase: correlations with acute graft-versus- F, Masera G, Barbui F, Pelicci PG. Molecular evaluation of host disease and relapse. Br J Haematol. 1993; 84: 67–74. residual disease as a predictor of relapse in acute promyelocytic 21 Lanotte M, Martin-Thouvenin V, Najman S, Ballerini P, Valensi F, leukemia. Lancet 1992; 340: 1437–1438. Berger R. NB4, a maturation inducible cell line with t(15;17) 8 Miller WH Jr, Levine K, DeBlasio A, Frankel SR, Dmitrovsky E, marker isolated from a human acute promyelocytic leukemia Warrell JR. Detection of minimal residual disease in acute promye- (M3). Blood 1991; 75:1080–1086. locytic leukemia by reverse transcriptase polymerase chain reac- 22 Tobal K, Liu Yin JA. RT-PCR method with increased sensitivity tion assay for PML-RARα fusion mRNA. Blood 1993; 82: 1689– shows persistence of PML-RARA fusion transcripts in patients in 1694. long-term remission of APL. Leukemia 1998; 12: 1349–1354. 9 Ikeda K, Sasaki K, Tasaka T, Nagai M, Kawanishi K, Takahara J, 23 Diverio D, Rossi V, Avvisati G, De Santis S, Pistilli A, Pane F, Irino S. Reverse transcription-polymerase chain reaction for PML- Saglio G, Martinelli G, Petti MC, Santoro A, Pelicci PG, Mandelli RAR alpha fusion transcripts in acute promyelocytic leukemia and F, Biondi A, Lo Coco FL. Early detection of relapse by prospective its application to minimal residual leukemia detection. Leukemia reverse transcriptase-polymerase chain reaction analysis of the 1993; 7: 544–548. PML/RARα fusion gene in patients with acute promyelocytic leu- 10 Huang W, Sun G-L, Li X-S, Cao Q, Lu Y, Jang G-S, Zhang F-Q, kemia enrolled in the GIMEMA-AIEOP multicenter ‘AIDA’ trial. Chai J-R, Wang Z-Y, Waxman S, Chen Z, Chen S-J. Acute promye- Blood 1998; 92: 784–789. locytic leukemia: clinical relevance of two major PML-RARα iso- 24 Burnett AK, Grimwade D, Solomon E, Wheatley K, Goldstone AH. forms and detection of minimal residual disease by retrotranscrip- Presenting white blood cell count and kinetics of molecular tase. Blood 1993; 82: 1264–1269. remission predict prognosis in acute promyelocytic leukemia 11 Laczika K, Mitterbauer G, Korninger L, Knobl P, Schwarzinger I, treated with all-trans retinoic acid: result of the randomized MRC Kapiotis S, Haas OA, Kyrle PA, Pont J, Oehler L, Purtscher B, Thal- trial. Blood 1999; 93: 4131–4143. hammer F, Lechner K, Jaeger U. Rapid achievement of PML-RAR 25 Lo Coco F, Diverio D, Avvisati G, Petti MC, Meloni G, Pogliani alpha polymerase chain reaction (PCR) negativity by combined EM, Biondi A, Rossi G, Carlo-Stella C, Selleri C, Martino B, Spec- treatment with all-trans retinoic acid and chemotherapy in acute chia G, Mandelli F. Therapy of molecular relapse in acute promye- promyelocytic leukemia: a pilot study. Leukemia 1994; 8: 1–5. locytic leukemia. Blood 1999; 94: 2225–2229.

Leukemia