Indian Journal of Cancer Medknow Publications on behalf of Indian Cancer Society ISSN: 0019-509X EISSN: 1998-4774 Vol. 48, Num. 1, 2011, pp. 60-67

Indian Journal of Cancer, Vol. 48, No. 1, January-March, 2011, pp. 60-67

Original Article

Role of RT-PCR and FISH in diagnosis and monitoring of acute promyelocytic leukemia

S Polampalli¹, A Choughule², K Prabhash¹, P Amare³, C Baisane³, S Kabre³, S Mahadik⁴, S Shinde⁴, R Nair¹, S Banavali¹

¹ Department of Medical Oncology, Tata Memorial Hospital, Mumbai, India
² Department of Medical Oncology and Hematopathology, Tata Memorial Hospital, Mumbai, India
³ Department of Cancer Cytogenetics, Tata Memorial Hospital, Mumbai, India
⁴ Department of Hematopathology, Tata Memorial Hospital, Mumbai, India

Correspondence Address:
K Prabhash
Department of Medical Oncology, Tata Memorial Hospital, Mumbai
India
kp_prabhash@rediffmail.com

Code Number: cn11010

Abstract

Keywords: Reverse transcriptase polymerase chain reaction, fluorescent in situ hybridization, acute promyelocytic leukemia

Introduction

Acute promyelocytic leukemia (APL) is characterized by the reciprocal translocation t(15;17)(q22;q21) leading to the formation of PML-RARA and RARA-PML fusion genes. ^[1],[2] PML-RARA is believed to be critical for leukemogenesis. ^[2],[3] In patients with morphological APL the presence of a PML-RARA rearrangement predicts a favorable response to retinoids such as ATRA, ^[4] which, in combination with chemotherapy, has been found to provide significant improvement in survival. ^[5] Therefore, establishing the presence of a PML-RARA rearrangement in suspected APL cases is valuable for optimal patient management, and for the analysis of clinical trials. Generally, conventional karyotyping (CK), fluorescence in situ hybridization (FISH), and reverse transcriptase-polymerase chain reaction (RT-PCR) are the techniques used for detecting this abnormality. Early studies in specialized centers suggested that t(15;17) can be detected by cytogenetics in virtually all cases of morphological APL. ^[6] CK has been a gold standard for the detection of unique translocation markers and additional chromosomal aberrations such as trisomy and tetrasomy 8, iso 17q, and other variant translocations such as t(15;19;17). ^[7],[8],[9] However, large multi-center clinical trials, such as the MRC ATRA trial, have demonstrated that t(15;17) is not detected by cytogenetics in at least 10% of the cases in which PML-RARA rearrangements are demonstrated by molecular methods such as FISH and RT-PCR. ^[10],[11] In recent studies, FISH on metaphases and interphases from cytogenetic preparations has been shown to be a useful tool in identifying PML-RARA in APL patients at the time of diagnosis and for minimal residual disease. ^[12],[13] The RT-PCR status after completion of therapy is of prognostic significance and emphasizes the importance of the achievement of molecular remission as a prerequisite for long-term survival. RT-PCR allows the identification of one abnormal cell out of 10 ⁴ -10 ⁶ . ^[14],[15] FISH on the other hand has a detection limit of 10 ^-3 in treated patients. ^[12] Most of these studies are individual studies, either focusing on RT-PCR or FISH. However, there have not been many studies, except for few, ^[12] where both these techniques were compared together, in order to find the advantages and limitations of one technique in comparison to another technique.

The objective of this study was to compare and assess the role of FISH and RT-PCR in the diagnosis and long-term monitoring of APL.

Materials and Methods

We compared 52 diagnostic samples received for RT-PCR and FISH at baseline during the period 2006 - 2008. We also compared the follow-up samples received for RT-PCR and FISH, for minimal residual monitoring after initiation of therapy, at different interval of times, to analyze the sensitivity of both the techniques. The protocols used for both the techniques were as follows;

Reverse transcriptase-polymerase chain reaction

Total RNA was extracted from the bone marrow samples using the standard techniques. cDNA was prepared by using the commercially available kit (Fermentas), using Random Hexamer primer. To develop the sensitivity and specificity for PML-RARA, HOT start RT-PCR was performed, ^[16] using a specific primer for PML-RARA, with forward primer: 5′ACC GAT GGC TTC GAC GAG TTC3′; and reverse primer: 5′AGC CCT TGC AGC CCT CAC AG 3′. PCR conditions were as follows: Initial denaturation at 95oC for 15 minutes, followed by five cycles, each consisting of denaturation at 96oC for one minute, annealing at 55oC for 15 minutes, extension at 72oC for two minutes, and final extension at 72oC for two minutes. This was followed by 30 cycles, each consisting of denaturation at 94oC for 35 seconds, annealing at 55oC for one minute, extension at 72oC for three minutes, and final extension at 72oC for 10 minutes. Final cooling was at 4oC. The PCR product was electrophoresed on 2% agarose gel and the band was analyzed. The long (L) form showed three bands at 291, 550, and 691 base pairs; while the short (S) form showed a single band at the 220 base pair. With each batch of samples, positive and negative controls were run. For internal controls, β-actin mRNA was used.

Fluorescence in situ hybridization

Fluorescence in situ hybridization (FISH) was carried out using the direct fluorescence method, on fixed cytogenetic specimens of bone marrow aspirate samples as per protocols supplied by Vysis Inc / Oncor, with some modifications. ^[17],[18] After dehydrating the slides in ethanol grades (70, 85, and 100%), slides were co-denatured along with a FISH probe on a Hybrite machine for five minutes at 73ºC, followed by overnight hybridization at 37ºC. Stringent washes of the slides were done in 0.4XSSCN (NP 40 - 0.3%) at 73 ° C for two minutes followed by 2XSSCN (NP 40 - 0.1%) at room temperature for one minute. DAPI / Antifade was applied and signals were observed under an epifluorescent microscope. FISH studies were performed using the LSI dual fusion dual color probe for PML and RARa (D-FISH) (Abott Molecular-Vysis, Des Plaines, IL; Delkenheim, Germany), which allows identification of reciprocal PML-RARA and RARA-PML and also additional aberrations. The RARA break-apart probe (Abott Molecular-Vysis, Des Plaines, IL; Delkenheim, Germany) was used to increase the diagnostic sensitivity, which could simultaneously detect variant translocations such as t(11;17) and t(5;17), along with t(15,17). Two hundred interphases cells were counted by two observers. The cut-off value was taken as 10% by simultaneously studying the normal samples. Positive and negative controls were used with each batch.

Results

Baseline results

Out of 52 baseline samples which were diagnosed as APL, 50 samples were diagnosed by RT-PCR and 48 samples by FISH [Table - 1].

Out of a total of six negative samples, four samples were positive by RT-PCR and two positive by FISH, with failure rate for RT-PCR being 4% and that for FISH being 8%.

Two samples had very faint RNA, which led to false negative results by RT-PCR.

Minimal residual disease monitoring

After initiation of treatment with all-trans retinoic acid (ATRA) along with chemotherapy, the samples were sent for RT-PCR and FISH studies for minimal residual monitoring at the end of induction, at the end of consolidation, and at different intervals of maintenance therapy. We analyzed these samples to assess the sensitivity of both the techniques at the end of induction, consolidation, and maintenance.

Samples analyzed at the end of induction

(First follow-up)

Forty-two samples were received for RT-PCR monitoring of minimal residual disease monitoring at the end of induction. These samples were compared with the FISH results to assess the sensitivity of both the techniques. Out of the 42 samples, two samples did not have any RNA, and therefore, were not processed further. However, these two samples were negative by FISH. The remaining 40 samples were compared by both the techniques as shown in [Table - 2].

Out of the 40 samples, eight samples (20%) showed the presence of PML-RARA by RT-PCR and / or FISH. Discordance between RT-PCR and FISH was seen in four samples. Out of these four samples, one sample was negative by RT-PCR, but positive by FISH, with a failure rate of 13% (failed to detect PML-RARA in one out of eight positive samples). Although, on the other hand, three samples were negative by FISH, but positive by RT-PCR, with a failure rate of 38% (failed to detect PML-RARA in three out of eight positive samples).

Samples analyzed at the end of consolidation (Second follow-up)

Twenty-six samples were analyzed at the end of consolidation by RT-PCR and FISH. Out of these 26 samples, three samples (12%) were positive by RT-PCR and / or FISH [Table - 3]. Discordance was seen in two samples, which were positive by RT-PCR, but negative by FISH, with the failure rate for FISH being 66% (failed to detect PML-RARA in two out of three positive samples).

Out of these three cases, two cases were positive by RT-PCR, but negative by FISH in the previous follow-up sample, that is, at the end of induction, while the third sample was negative, both by RT-PCR and FISH in the previous follow-up.

Samples analyzed during maintenance

Twenty-seven samples were analyzed by RT-PCR and FISH during maintenance, out of which four samples showed the presence of PML-RARA, both by RT-PCR and FISH, while the rest of the samples were negative by both the techniques.

Significance of RT-PCR and FISH results in monitoring MRD

We tried to correlate the results of RT-PCR and FISH to find out their significance in the monitoring of MRD in APL patients. It was seen that all the cases that were negative by RT-PCR and FISH at end of induction and consolidation, continued to remain in molecular remission with a median follow-up of 18 months.. There were four patients (cases 1 to 4, [Table - 4]) who were positive both by RT-PCR and FISH at the end of induction, out of which two patients did not achieve molecular remission, while two patients were lost to follow-up. Out of the two patients who did not achieve molecular remission, one patient is alive while the other patient expired. There were three patients (cases 5 to 7) who were positive only by RT-PCR, but not by FISH. Out of these three patients, one patient did not achieve molecular remission and subsequently expired, even as the other two patients were in molecular remission, with a median follow-up of 16 months. There was one patient (case 8) who was negative by RT-PCR, but positive by FISH. This patient continues to be in molecular remission with a median follow-up of 12 months.

Significance of MRD by RT-PCR and FISH at the end of induction

Out of the seven patients who were positive by RT-PCR at the end of induction, three patients relapsed, two were in remission, and two were lost to follow-up. The predictive value of relapse in the patients who were positive by RT-PCR at end of induction was 60 ± 11%. However, out of 33 patients who were negative at the end of induction by RT-PCR, only one patient relapsed, while all the rest continued to be in remission. The predictive value of relapse in the patients who were negative by RT-PCR at the end of induction was 3%. On the other hand, out of five patients who were positive by FISH at the end of induction, two patients relapsed, one patient was in remission, and two patients were lost to follow-up. Thus, the predictive value of relapse in the patients who were positive by FISH at the end of induction was 57 ± 13%. Out of 35 patients who were negative by FISH at the end of induction, two patients relapsed, while all the rest continued to be in remission. The predictive value of relapse in the patients who were negative by FISH at end of induction was 6%.

Significance of minimal residual disease by transcriptase polymerase chain reaction and fluorescent in situ hybridization at the end of consolidation

Out of the three patients who were positive by RT-PCR at end of consolidation, two patients relapsed while one patient was in remission. The predictive value of relapse in the patients who were positive by RT-PCR at the end of consolidation was 67%. Out of the other 23 patients who were negative by RT-PCR at end of consolidation, one patient relapsed, while all the rest were in molecular remission. Thus the predictive value of relapse in the patients who were negative by RT-PCR at the end of consolidation was 4%. On the other hand, all patients except one were negative by FISH at the end of consolidation. Out of these 22 patients, three patients relapsed, while all the rest were in remission. Thus, the predictive value of relapse in the patients who were negative by FISH at the end of consolidation was 14%.

Discussion

Acute promyelocytic leukemia (APL or AML-M3) is a subtype of acute myeloid leukemia with distinct clinical and morphological features. Historically one of the most lethal forms of acute myeloid leukemia, APL leads to disseminated intravascular coagulation and death when not diagnosed and treated. Treatment with all-trans -retinoic acid substantially improves survival in patients who have failed anthracycline chemotherapy or for whom anthracycline is contraindicated. Similarly, arsenic trioxide is beneficial in APL patients who have failed or have contraindications for treatment with anthracycline or retinoid-based therapy. Genetically, APL is characterized by a unique chromosomal anomaly. More than 99% of the APL patients harbor a translocation between chromosomes 15 and 17, which fuse the retinoic acid receptor alpha (RARA) gene on chromosome 17 with the PML gene on chromosome 15. Detection of the PML-RARA t (15;17) translocation is diagnostic for APL, although the diagnosis can also be based on morphology. The presence of this translocation is necessary for response to the all-trans -retinoic acid and arsenic trioxide. Thus, the PML-RARA t (15;17) assay is useful for diagnosis and predicting treatment response. It is also helpful for monitoring the therapeutic response and MRD and for detecting early relapse. Both FISH and PCR methods may be used in the diagnosis of APL, to monitor patients for APL recurrence, and to predict a therapeutic response to all-trans -retinoic acid and arsenic trioxide therapies. The application of FISH not only allows to detect the reciprocal fusion of PML-RARA, but also the additional aberrations of 17q deletion, residual deletion of PML (unpublished data), and iso 17q. ^[8] These advantages of FISH technology at diagnosis are unique and cannot be detected by any other molecular technology including RT-PCR. Although, on the other hand, for suspected low-level, recurrent or MRD detection and monitoring, RT-PCR is recommended. The objective of this study is to compare and assess the role and sensitivity of FISH and RT-PCR in the diagnosis and monitoring of minimal residual disease in APL cases.

In the present study, the failure rate for RT-PCR at baseline was 4% and that for FISH was 8%.Out of the 52 samples received for RT-PCR, two samples had very faint RNA, which led to false negative results. In one of the studies conducted by S Iqbal et al. , ^[19] out of the 28 APL cases, none were negative by RT-PCR, but three samples were negative by FISH, with a failure rate of 11%. Both the techniques had certain limitations in diagnosing APL. Absence of the t (15;17) by CK in APL cases most commonly reflected poor quality metaphases or cryptic rearrangements due to insertion events. FISH also failed to detect these cryptic insertions if locus-specific probes were not used. ^{[3],[20],[21]} however, these cryptic insertions could be easily detected by simple RT-PCR techniques. ^{[3],[20],[21]} According to most investigators, high-quality RNA and efficient RT were the crucial determinants for successful RT-PCR of PML-RARA ^{[14],[22],[23]} On account of frequent leukopenia and the associated coagulopathy, the yield and quality of RNA from diagnostic samples were frequently poor, and this led to failure and false negativity by RT-PCR. This required a repeat sample, which in turn added to the sample load of the laboratory and also caused inconvenience to the patient, as the patient had to undergo repeat bone marrow aspiration. In such cases, FISH on interphase nuclei was a reliable tool for the diagnosis of APL, with a sensitivity greater than that of FISH on metaphase cells and super imposable to that of RT-PCR. ^[19]

The presence of a specific and detectable tumor marker in leukemic cells allows investigators to molecularly assess the response to therapy and MRD in every patient with APL. Both FISH and RT-PCR are being used to molecularly monitor these patients. Although, a number of studies employing conventional PML-RARA assays, which typically detect up to one APL cell in 10 ⁴ non-leukemic cells, have shown that extended courses of ATRA and combination chemotherapy ultimately induce PCR negativity in a vast majority of patients. ^{[11],[25],[26],[27],[28],[29]}

We analyzed the sensitivity of both these techniques for monitoring the minimal residual disease at the end of induction and consolidation. There have not been many studies, except for a few, ^[12],[30] which have conducted a comparison between RT-PCR and FISH, to monitor MRD in APL patients. In one of the studies conducted by Mancini M et al. , ^[12] 11 patients who were investigated during complete remission (CR) by both FISH and RT-PCR, in order to evaluate residual disease; showed a very good correlation between RT-PCR and FISH. Although, in another study conducted by MLLF. Chauffaille, ^[30] of the six samples evaluated after treatment, the failure rate for FISH to detect residual disease was almost 50%. In the present study, discordance between RT-PCR and FISH was seen in four samples, at the end of induction. The failure rate for RT-PCR was 13% and that for FISH was 38%. Three samples were positive at the end of consolidation, with RT-PCR being positive in all the three cases, while FISH was positive in only two cases, with the failure rate being 66%. Four patients relapsed during maintenance; they were detected by both RT-PCR and FISH, which showed a good correlation Thus, it shows that the sensitivity of RT-PCR is more as compared to FISH, in detecting minimal residual disease up to the phase of consolidation, which is very critical for patient management and treatment. However, quality of the RNA has to be checked before, as this can lead to false negative results as seen in our study.

We further analyzed the significance of these MRD results obtained by both the techniques with respect to the remission status of the patients, in order to assess the false positive or negative results. It was seen that the predictive value of relapse in the patients who were positive by RT-PCR at end of induction was 60 ± 11%, and for those who were negative by RT-PCR, it was 3%. On the other hand the predictive value of relapse in the patients who were positive by FISH at end of induction was 57 ± 13%, and for those who were negative, it was 6%. Similarly, the predictive value of relapse in the patients who were positive by RT-PCR at end of consolidation was 67%, and for those who were negative, it was 4%. On the other hand the predictive value of relapse in the patients who were negative by FISH at the end of consolidation was 14%.

Till date there has not been any study that has compared both these techniques to assess the significance of MRD results obtained by both these techniques, with the remission status. The present study shows that the predictive value of relapse is associated more with the MRD results by RT-PCR than that by FISH.

Achievement of molecular remission is an important therapeutic goal in APL; as persistence of PML-RARA fusion transcripts until the end of chemotherapy consolidation, which occurs in 2 - 8% of the patients, ^{[11],[26],[31],[32]} has been found to be highly predictive of relapse, ^[33] which can, however, be averted by additional therapy such as allogenic BMT. ^[34] Even the present study shows that the positive results obtained by RT-PCR at the end of consolidation were more associated with relapse than with the results obtained at the end of induction, (60% at the end of induction, 67% at the end of consolidation). However, it was also seen that the some of the patients who were negative at the end of consolidation, eventually relapsed (predictive value being 4%). There have been studies that have shown that patients who are negative by RT-PCR ultimately go on to relapse. ^{[11],[26],[33],[35],[36]} This indicates that PCR assessment at this single time-point cannot be relied upon to determine the optimum treatment approach in individual patients. In the study reported by the GIMEMA group, which employed the AIDA protocol, no correlation was observed between the PCR status after induction and subsequent relapse risk. ^[26] This was consistent with the concept that detection of PML-RARA transcripts at this stage could very well relate to differentiating leukemic cells, subject to subsequent apoptosis, which could not be distinguished by conventional RT-PCR protocols from the residual APL blasts. In a recent study the Italian GIMEMA group reported that recurrence of PCR positivity detected by three monthly surveillance marrows performed after completion of therapy, was highly predictive of relapse. ^[33] By using such a strategy approximately 70% of the relapses were successfully predicted, with the majority (81%) of patients who ultimately relapsed converting to PCR positivity, within the first six months following completion of therapy. Median time from detection of molecular relapse to hematological relapse was three months, with a range of 1 - 14 months. However, for the MRC study in which the PCR status used PML-RARA and RARA-PML assays, it was determined after each course of chemotherapy, which suggested that the rate of clearance of disease-related transcripts was indeed an independent prognostic factor in APL. Detection of transcripts at any stage following induction or during consolidation therapy was associated with an increased risk of relapse. Therefore, while it was clear that reduction of PML-RARA transcripts to below the detectable threshold by standard RT-PCR assays (one in 10 ⁴ ) was a prerequisite for long-term survival, and could currently represent our best therapeutic goal, achievement of PCR negativity could not be equated with cure.

The failure to detect residual disease in the marrow at the end of consolidation therapy in a significant proportion of patients, who ultimately relapsed, has highlighted the relative insensitivity of such assays, which has been ascribed to inefficiency of the RT step, and also to the relatively low expression of PML-RARA. ^[22] Indeed, analyses based on RNA derived from peripheral blood have been found to be even less sensitive, ^[37] thereby accounting for the reliance upon bone marrow sampling for MRD detection in this disease, to date. However, it also remains a possibility that in many cases, failure to detect residual disease is a reflection of sample quality giving rise to ′false-negative′ results; while very occasionally MRD analyses performed on the marrow will fail to predict isolated extramedullary relapses. ^[30] In view of the relative insensitivity of the conventional nested PML-RARA RT-PCR assay, a number of different approaches such as, hotstart ^[23] PCR and real time PCR ^[37] have been used, to attempt to more accurately identify a subgroup of patients at a particularly high-risk of subsequent relapse. The cost of Real time PCR, although more sensitive than the conventional PCR, limits its usage for regular monitoring of APL patients. In order to circumvent this problem, we have been using the hotstart PCR along with recombinant Taq polymerase (HOTstar, Qiagen), which has increased the efficiency of RT-PCR and increased the sensitivity threshold of the detection of minimal residual in APL cases.

Conclusions

Both RT-PCR and FISH are important for diagnosis of APL cases, as both the techniques complement each other in the absence or failure of any of them. In order to provide an accurate diagnosis, CK, FISH, and RT-PCR are mandatory in all newly suspected cases of APL. Despite its limitations, CK should continue to be used in diagnosis for the detection of karyotype patterns involving other chromosomes besides t (15;17). FISH offers some unique advantages over RT-PCR, where a single probe covers multiple dispersed breakpoints in chromosomal regions and can detect additional aberrations. Although we do not know the clinical significance of these aberrations, the long-term follow-up of such cases, in a large series, will help focus their clinical significance. Also with the use of cosmid probes, rare events of cryptic translocations can also be detected, which otherwise might be missed by normal probes, but can be detected by RT-PCR. RT-PCR is more sensitive than FISH for the detection of minimal residual disease, in long-term monitoring of these patients. Also, the results of the predictive value of relapse correlate more with RT-PCR than FISH. However, further investigation is needed to better establish the clinical value of molecular monitoring. In particular, the following have to be clarified: (1) whether a greater sensitivity of the RT-PCR assay for PML-RARA will allow better identification of patients at risk of relapse at the end of consolidation; (2) if newer quantitative technologies may provide earlier monitoring that is clinically significant; and (3) whether anticipation of salvage therapy in patients treated for molecular recurrence is more advantageous than the treatment of a hematological relapse. As to this latter issue, studies are underway, ^[38] but it may be anticipated that the therapy of molecular relapse can at least minimize the significant mortality rate observed during re-induction of overt disease.

References

Copyright 2011 - Indian Journal of Cancer

The following images related to this document are available:

Photo images