Annals of Oncology 1: 167-170.1990. © 1990 KUtwer Academic Publishers. Printed in the Netherlands.
Special article Minimal residual disease in leukaemia and lymphoma RE. Cotter, C. Price, B.D. Young & T.A. Lister ICRF Department of Medical Oncology, St. Bartholomew's Hospital, London, UK
Introduction The leukaemias and lymphomas may be contrasted sharply with most other malignancies by the fact that, despite wide spread dissemination at presentation many patients may be returned to normality, sometimes permanently following therapy. Survival correlates closely with the extent to which the disease has been eliminated, or the completeness of remission. It is of considerable practical importance to define the response to therapy as precisely as possible. Thus any technique permitting the detection of disease at a lower level than conventionally possible might be of enormous benefit, possibly allowing modification of therapy to convert a 'complete remission' destined to relapse (the majority) into cure, or alternatively the termination of therapy in a patient already cured. Minimal Residual Disease (MRD) is a clumsy and contentious term, by definition describing the lowest level of disease detectable with the methods available. It is however in common use. The most sensitive techniques are often able to demonstrate molecular and cellular change in the neoplastic cells, in morphologically and cytologically normal tissue. The determination of tumour specific markers [1-3] is important in order to devise a strategy for detection of MRD. There is a need for rapidity, ease and certainty of results and techniques involving enzymatic amplification (PCR)  offer the best opportunity of attaining these aims. Southern analysis allows similar specificity however there is a much lengthier procedure with a marked loss of sensitivity [5-7].
Methods to detect MRD Potymerase chain reaction (PCR). The ability to exponentially amplify a target sequence of DNA in vitro by enzymatic synthesis (PCR) was first described in 1985 , Two synthetically prepared oligonucleotides primers hybridise to opposite strands flanking the region of interest in the target DNA. The intervening sequences are generated by extension from these primers by a DNA polymerase, thus duplicating the target DNA. There is theoretical doubling of DNA yield for every cycle leading to an amplification of the target DNA by a million-fold (220) over 20 cycles, a level detectable by gel electrophoresis or Southern
analysis. The application of PCR to the detection of minimal residual disease requires the target DNA to be a marker from the tumour cells. This is illustrated well by the t(14; 18) translocation associated with 85% of follicular nonHodgkins lymphomas (NHL) and 20% of high grade diffuse NHL [8-11]. The translocation involves the Bcl-2 gene at one of two sites (the major breakpoint region and minor cluster region) on chromosome 18 and the immunoglobulin heavy chain joining region on chromosome 14. The breakpoints cluster within a few hundred base pairs on both chromosomes enabling detection of disease by this technique and has the sensitivity to detecting at least one tumour cell in 105 or 106 cells . The sensitivity may be increased by either enrichment of the sample for tumour cells by the use of flow cytometry followed by PCR or by a 'booster' PCR technique  or utilising two sets of primers flanking the target DNA. In contrast many tumours show alterations over much longer segments of DNA, sometimes over hundreds of kilobases [13-15]. However the length of DNA sequence may consist of mainly intron sequence that is not expressed in the messenger RNA (mRNA) transcript This mRNA may be short enough to facilitate the use of the PCR technique to look for alterations over a considerable length within the transcribed exon DNA. Abnormal expression of mRNA by tumour cells acts as a marker of disease . This is demonstrated well in chronic myeloid leukaemia (CML), some forms of acute lymphocytic leukaemia (ALL) and occassionally in acute myeloid leukaemia (AML) where a chimaeric mRNA (BCR-ABL), from the Philadelphia chromosome, is found exclusively in the leukaemic cells [13, 15]. The BCR exons involved in the translocation may be separated by over a hundred kilobase of intron sequence [13, 15]. Utilising a modified PCR technique it is possible to detect this abnormal mRNA marker of the disease. Complimentary DNA (cDNA) from the chimaeric mRNA is produced by an initial step involving a reverse transcriptase enzyme, followed by the PCR procedure for DNA. Resultant amplification of the cDNA occurs [13-15]. Amplification from the mRNA is advantageous in that interference from intron sequence does not occur and therefore analysis of changes within longer sections of DNA is possible [13-15]. PCR is most suited for detection of target DNA of less than a 1.000 nucleotide base pairs . The quantity of
168 DNA required is extremely small , 100 times less than for Southern analysis, while giving a least a 103 times greater sensitivity [6-10, 16]. In addition PCR has been applied to small lengths of target DNA allowing the use of this method on degraded DNA, unsuitable for Southern analysis, such as paraffin embedded sections . Retrospective examination of archival material for tumour markers and clonality is thus facilitated for prospective follow up of patients with MRD. Sequencing reactions readily carried out directly on the PCR products allows the tumour specific alterations to be analysed down to the single nucleotide base level. In this way, sequence analysis from tumour material gives a very clear and precise clonal marker for the patients disease  and may be helpful in determining if recurrent disease shows the same clonality as the original tumour. In some cases it may be possible to use the PCR product as a unique marker for the patients tumour as shown with amplification of the T cell receptor in T cell lymphoid malignancies [18, 19]. The use of an immunoglobulin heavy chain variable region consensus sequence in combination with a heavy chain joining sequence has broadened the application of PCR to the majority of B cell lymphoid malignancies . The detection of minimal residual disease by PCR has the advantage of rapidity (approximately 3 hours per reaction) greatly increased sensitivity over other techniques and therefore the ability to detect sub-clinical disease at a much earlier stage [8, 13]. The main disadvantage lies in the necessity for the neoplastic disease to have identified flanking sequences of DNA or mRNA that will allow amplification of a tumour specific marker. PCR is an extremely sensitive method for amplification of a target DNA [4, 8, 13], however is not directly quantitative. Adaptation of PCR to allow quantitation for detection of proviral DNA has been demonstrated . Application of quantitation techniques to MRD could be of benefit in sequential follow up, charting progression of disease from an early sub-clinical stage. Further clarification of tumour specific translocations, deletions and alterations will permit greater application of this technique to ascertain the presence or absence of MRD. Alternative techniques. A number of methods are presently in use for the detection of MRD in addition to PCR, however none offer a similar degree of sensitivity. Southern analysis and immunophenotyping are techniques that discern disease at a level of 1 in 100 cells [6, 7, 22]. The sensitivity of Southern analysis has been improved by initial enrichment for tumour cells by flow cytometry improving the level of detection to 1 in 1.000 cells . Immunophenotyping is less specific than molecular analysis requiring therefore a battery of antibodies taking more time while sometimes causing difficulties with the interpretation [23, 24]. Recently a technique for rapid RNA transcript sequence analysis of the clonal T cell receptor or immunoglobulin heavy chain rearrangement from malignant BMT (bone marrow tumour) lymphocytes was described. This enables the generation of individual tumour probes in lymphomas and leukemias to facilitate tissue in situ analysis
for MRD . However, this technique could be further developed to increase sensitivity of detection of MRD by incorporating the sequence data obtained into the PCR setting. Other techniques such as digital image microscopy for rare event detection using fluorescence probes have been developed and are reported to detect 1 in 10.000 tumour cells in bone marrow patients with leukemia but as yet arc not generally applicable .
Discussion The overall clinical relevance to patients with leukaemias and lymphoma of laboratory techniques able to detect minute quantities of tumour associated molecular abnormalities will depend on several factors, excluding the unanswered questions relating to false negative or positive results (the sensitivity of the techniques may allow contamination to contribute to rogue results). Although non random chromosomal changes have been reported for an increasing number of haematological malignancies [1-3, 6], the use of molecular probes for Southern analysis  and subsequent development of PCR methods to detect these alterations [4, 8, 13] has only been applicable for the bcr gene in chronic myeloid leukaemia [13-15], the bcl-2 gene in follicular lymphoma [8-11] and the immunoglobulin heavy chain  and T cell receptor genes in lymphoid malignancies [18, 19]. This clearly limits the applicability of the techniques to a relatively modest number of groups of patients at present, if however in no way limits the investigation of the hypothesis that 'molecular evidence of disease' after treatment equates with a inevitability of recurrence occurring sooner or later. Monitoring of remission, with other, less sensitive and probably less 'disease specific' methods has only met with limited success [6, 7, 16, 22-24]. Follow up of patient in clinical and haematological remission of acute lymphoblastic leukemic using anti CALLA antibody to examine the marrow known to have been positive at presentation failed to be of use due to phenotype shift at relapse . However a study, in which two immunological parameters, tailored to the original phenotype of the blast cells were utilised, produced results highly predictive of relapse . Investigation of patients presenting with different stages of lymphoma or leukaemia using flow cytofluormetric analysis for K or \ light chain expression , or gene rearrangement analysis suggests that clonal evidence of disease may be found in the absence of clinical or morphological findings [6, 22]. Studies of the peripheral blood of patients in long term follow up of malignant lymphoma using RFLP (restrictive fragment length polymorphisms) or PCR for the t(14; 18) translocation show persistant abnormalities in a proportion despite continuing clinical remission [8, 30, 31]. Clearly if the abnormalities are present for many years without the evidence of recurrence, their clinical relevance must be queried . Satisfactory methods for quantifying the amount of abnormal circulating DNA [21, 33] will obviously offer the opportunity to look for progression or regression over time.
169 Such techniques may have an application for determining the efficacy of in vitro treatment of bone marrow for autologous transplantation . Much controversy surround the necessity or reliability of 'clean up' and is hampered by the difficulty of proving that the procedure either does anything, or contributes to outcome. Preliminary data using the PCR with probes for the bcl-2, suggest that in vitro treatment of bone marrow from patients with follicular lymphoma with anti-B-1 and complement depletes it by 10-50 fold, but rarely cleans it completely . Similar results have been shown with cell lines or multiple antibodies . Ultimately these techniques will only be of major clinical relevance when treatment is available which can convert a clinical but not molecular remission into a clinical and 'molecular' complete remission and thereby cure. In addition, it must be borne in mind that most cytotoxic therapies are dangerous, and the confirmation of remission at the molecular level might allow some patients to stop therapy sooner. Many questions remain to be answered about 'minimal residual disease'. Indeed, this is a relatively small problem in haematological malignancy. A far greater problem concerns the inability of treatment to achieve even clinical remission for most patients. Under no circumstances must the investigation of molecular generation of leukaemia and lymphoma be inhibited by this. On the contrary it must be persued along the current lines with vigour and enthusiasm. Greater understanding of the diseases continues to evolve from the laboratory and the fact that therapeutic advances are lagging behind should only serve as a stimulus for forward thinking.
References 1. Sampi K. Significance of tumor markers in the treatment of cancer hematologic malignancy. Gan-To-Kagaku-Ryoho 1987; 14: 3041-5. 2. Collins SJ. Direct sequencing of amplified genomic fragments documents N-ras point mutations in myeloid leukemia. Oncogene-Res 1988:3:117-23. 3. Lyons J, Janssen JW, Bartram C, Layton M, Mufti GJ. Mutation of Ki-ras and N-ras oncogenes in myelodysplastic syndromes. Blood 1988; 71:1707-12. 4. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn OT, Mullis KB, Eriich HA. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988; 239: 487-91. 5. Southern EM. Detection of specific sequences among DNA fragments separated by gel electrophoresi; J. Mol Biol 1975; 98: 503-17. 6. Katz F, Ball L, Gibbons B, Chessells J. The use of DNA probes to monitor minimal residual disease in childhood acute lymphoblastic leukaemia. Br J Cancer 1989; 73: 173-80. 7. Henni T, Vidaud M, Bretagne S, Goosens M. Detection of residual disease in onco-hematology: the contribution of molecular biology. Pathol Biol Paris 1988; 36: 91-5. 8. Crescenzi M, Seto M, Herzig OP, Weiss PD, Griffith RC, Korsmeyer SJ. Thermostable DNA polymerase chain amplification of 1(14; 18) chromosome breakpoints and detection of minimal residual disease. Proc Nail Acad Sci USA 1988; 85:4869-73. 9. Lee MS, Chang KS, Cabanillas F. Freireich EJ, Trujillo JM, Stan SA. Detection of minimal residual cells carrying the t(14; 18) by
DNA sequence amplification. Science 1987; 237: 175-8. 10. Ngan BY, Nourse J, deary ML. Detection of chromosomal translocation t(14; 18) within the minor cluster region of bcl-2 by polymerase chain reaction and direct genomic sequencing of the enzymatically amplified DNA in follicular lymphomas. Blood 1989; 73: 1759-62. 11. Weiss LM, Wamke RA, Sklar J, Deary ML. Molecular analysis of the t(14; 18) chromosomal translocation in malignant lymphomas. N Engl J Med 1987; 317: 1185-8. 12. Ruano G, Fenton W, Kidd KK. Biphasic amplification of very dilute DNA samples via 'booster'; PCR. Nucleic Add Research 1989; 17: 5407. 13. Lange W, Synder DS, Castro R, Rossi JJ, Blume KG. Detection by enzymatic amplification of bcr-abl mRNA in peripheral Mood and bone marrow cells of patients with chronic myelogenous leukemia. Blood 1989; 73: 1735-41. 14. Sarkar G, Sommer SS. Access to a messenger RNA sequence or its protein product is not limited by tissue or species specificity. Science 1989; 244: 331-4. 15. Evinger-Hodges MJ, Spinolo JA, Spencer V, Nieto P, Dicke KA. Detection of minimal residual disease in acute myelogenous leukemia by RNA in situ hybridization. Bone Marrow Transplant 1989; 4: 13-5. 16. Bregni M, Siena S, Neri A, Bassan R, Barbiri T, Delia D, Bonadonna G, Dalla-Favera R, Ganni AM. Minimal residual disease in acute lymphoblastic leukemia detected by immune selection and gene rearrangement analysis. J Clin Oncol 1989; 7: 338-43. 17. Price CG, Cotter FE, Curling OM, D'Ardenne AJ, Rohatiner AZS, Young BD, Lister TA. Polymerase chain reaction to confirm extranodal progression of follicular lymphoma. Lancet 1989; 1: 1132. 18. Hansen-Hagge TE, Yokota S, Bartram CR. Detection of minimal residual disease in acute lymphoblastic leukemia by in vitro amplification of rearranged T-ceU receptor S chain sequences. Blood 1989; 74: 1762-7. 19. d'Auriol L, Macintyre E, Galibert F, Sigaux F. In vitro amplification of T cell gamma gene rearrangements: a new tool for the assessment of minimal residual disease in acute lymphoblastic leukemias. Leukemia 1989; 3: 155-8. 20. McCarthy KP, Sloane JP, Wiedemann LM. A rapid PCR method for distinguishing clonal from polyclonal B-cell populations in surgical biopsies. Br J Cancer 1990 (in press). 21. Abbott MA, Poiesz BJ, Byrne BC, Kwok S, Sninsky JJ, Ehriich GD. Enzymatic gene amplification: qualitative and quantitative methods for detecting proviral DNA amplified in vitro. J-Infect-Dis 1988; 158:1158-69. 22. Brada M, Mizutani S, Molgaard H, Sloane JP, Treleaven J, Horwich A, Peckham MJ. Circulating lymphoma cells in patients with B and T non-Hodgkin's lymphoma detected by immunoglobulin and T-ceU receptor gene rearrangement. Br J Cancer 1987; 56: 147-52. 23. Bernard A. The limitations in utilizing phenotypic markers to detect minimal residual disease in acute leukaemias (ALL)-. Pathol Biol Paris 1988; 36: 17-20. 24. Visser, JW, Martens AC, Hagenbeek A. Detection of minimal residual disease in acute leukemia by flow cytometry. Ann NY Acad Sci 1986:468:268-75. 25. Seibel NL, Kirsch IR. Tumour detection through the use of immunoglobulin gene rearrangements combined with tissue in situ hybridization. Blood 1989; 74: 1791-5. 26. Lee BR, Haseman DB, Reynolds CP. A digital image microscopy system forrare-eventdetection using fluorescent probes. Cytometry 1989; 10: 256-62. 27. Greaves MF, Paxton AM, Jannossy G, Pain C, Johnson S, Lister TA. Acute lymphoblastic leukaemia associated antigen 111 alterations in expression during treatment and relapse. Leuk Res 1980; 4: 1. 28. Campana D, Coustan-Smith E, Jannossy G. Immunological methods of minimal disease detection. Blood 1990 (in press). 29. Smith BR, Weinberg DS, Robert NJ et al. Circulating monoclonal B lymphocytes in non-Hodgkin's lymphoma. N Engl J Med 1984; 311: 1476. 30. Fearon ER, Burke PJ, Scrriffer CA, Zehnbauger BA, Vogalstein B.
Differentiation of leukemic cells for polymorpholeucocytes in patients with ANLL. N EngI J Med 1986; 315: 15-24. Price CGA, Grant S, Dorey E, Cotter FE, Young BD, Rohatiner AZS, Lister TA. An increase in subclirdcal follicular lymphoma can be detected by PCR, and may be the best predictor of clinical relapse, (in press). Champlin R, Jacobs A, Gale RP, Boccia R, Elashoff R, Fbon K, Zieghelboim J. Prolonged survival in acute myelogenous leukaemia without maintenance chemotherapy. Lancet 1984a; 894-5. Price CGA, Cotter FE, Rohatiner AZS, Lister TA, Young BD. Estimation of number of cells bearing t(14; 18) in pathological samples from patients with follicular lymphoma Br J Cancer (in press). Rohatiner AZS, Barnett MJ, Amott S, Plowman N, Cotter F, Adams K, Batten EL, Outran S, Amess JAL, Horton MA, Lister TA. Ablative therapy supported by autologous bone marrow transplantation with in vitro treatment of marrow in patients with B-cell malignancy. Modem trends in human leukaemia VII. Edited by Neth, Gallo, Greaves and Kabisch. Springer-Verlag, Berlin, Heidelberg, 1987; 59-62.
35. Price CGA, Rohatiner AZS, Cotter FE, Watt S, Grant S, Young BD, Lister TA. Minimal residual disease and the efficacy of bone marrow purging in follicular lymphoma measured by an adapted PCR techniques. (Submitted for publication). 36. Negrin RS, Kiem HP, Schmidt-Wolf IGH, Blume KG, Cleary ML. Use of the polymerase chain reaction for monitoring the effectiveness of antibody plus compliment purging of bone marrow. Blood 1989; 74: 7 suppl 1; 241a. Received 9 February 1990; accepted 9 February 1990. Correspondence to: T.A. Liner, M.D. ICRF Department of Medical Oncology St. Bartholomew's Hospital London EC1A7BE, UK