British journal cJf Haenrutoloypj, 1986, 62. 385-391

The significance of host haemopoietic cells detected by cytogenetic analysis of bone marrow from recipients of bone marrow transplants

HELEN WALKER, C. K. J . SINGER, J A N E T PATTERSON,* A. H. GOLDSTONE A N D

H. G . PRENTICE* Depavtmcnt of Haernatology, University College Hospital, London, and *Academic Department of Haernatology, Royal Free Hospital, London

Received 14 November 1984; accepted for publication 2 3 May 19#5

SUMMARY. Bone marrow from 39 patients who received a bone marrow transplant (BMT) from a matched donor of different sex were studied by chromosome analysis for evidence of mixed haemopoietic chimaerism (MC). Recipient metaphases were detected in the bone marrow of 10 patients after BMT. Patients in whom MC was detected within 6 weeks of BMT did not all have a poor outcome. Two of seven are disease-free survivors at > 470 and > 632 d. All three patients in whom MC was diagnosed more than 6 weeks after BMT subsequently relapsed. Four factors appear to be important in determining the probability of relapse when MC is detected in a patient after BMT: the timing of detection of residual recipient cells; the proportion of these cells in the bone marrow: persistence of these cells in increasing proportions; and the karyotype of the recipient metaphases detected. Cytogenetic assessment may provide the earliest indication of relapse in these patients. In addition, this study provides further evidence that cyclophosphamide and total body irradiation, as used in these patients, may be inadequate conditioning therapy for BMT.

Recipients of bone marrow transplants (BMT) display haematological chimaerism (Thomas et a I , 1975: Sparkes et al , 1977). I n patients who have received a sex-mismatched BMT, the origin of bone marrow cells can be readily determined by identification of the sex chromosomes in metaphase preparations. Santos e t al (1972, 19 74), who studied patients conditioned with cyclophosphamide (Cy) alone, reported gradual disappearance of recipient marrow cells until 6 weeks, beyond which only donor metaphases were detectable. A similar pattern of engraftment has been reported in an animal model using immunoglobulin allotype markers (Santos & Tutschka. 1972). More recently Lawler et al (1984) studied patients conditioned with either c'y and total body irradiation (TBI) or melphalan. No recipient

Correspondence: Dr C. R. J. Singer, Department of Haematology. University College Hospital, Cower Street, London W C I E hAU.

38 5

386 Helen Walker et a1

metaphases were found in the blood or bone marrow during the first 9 weeks after BMT, though donor metaphases were found from the second post-transplant week. Subsequently recipient metaphases were detectable in marrow samples from seven of 39 patients. Three patients had a single recipient metaphase without subsequent relapse.

No clear relationship has yet been reported between mixed haemopoietic chimaerism (MC), i.e. detectable host metaphases in bone marrow after BMT and the risk of subsequent relapse. We therefore undertook this study to determine the incidence and significance of MC in a group of patients who received BMT conditioned with Cy and TBI.

PATIENTS AND METHODS

Thirty-nine patients who received an allogeneic BMT from a HLA-compatible donor of different sex between 1978 and 1983 as treatment for haematological malignancy were studied by the Cytogenetics Laboratory in the Department of Haematology, University College Hospital. Thirty-five patients were treated at the Koyal Free Hospital* and four were treated at University College Hospita1.j- Twenty-four patients were male, aged 6-43 years, and 1 5 were female, aged 5-36 years. The diagnosis was acute lymphoblastic leukaemia (ALL) in 2 1 patients (four first remission, 14 later remission and three relapse), acute non- lymphoblastic leukaemia (ANLL) in 14 patients (eight first remission, one later remission and five relapse), chronic phase of chronic granulocytic leukaemia (CGL) in one patient, and non- Hodgkin's lymphoma (NHL) in three patients (two first remission and one relapse). All but one patient were conditioned with cyclophosphamide (120 mg/kg) and single fraction TUI: 7.5 Gy, dose rate 1 5 cGy/min or 10 Gy, Dose rate 3.5 cGy/min (received mid-plane dose rates). One patient (AL32) was conditioned with melphalan (200 mg/m2). Blood products were irradiated to 1 5 Gy.

Direct bone marrow preparations for cytogenetic analysis were carried out using a modification of the method of Tjio & Whang (1962). Giemsa banding studies were performed using a modification of the method of Summer et al(1971). Slides were incubated at 60°C for 2 h in 2 x SCC, followed by rinsing in distilled water and Giemsa staining. The karyotypes were described according to standard international nomenclature (ISCN, 1978).

RESULTS

A mean of 35 metaphases were studied on bone marrow samples (rangc 4-88) from all 39 patients after BMT. Metaphase cells of recipient origin were identified after BMT in the bone marrow of 10 patients without evidence of relapse at the time of' cytogenetic analysis. The age-range and diagnoses of the patients in this group were similar to those of patients in whom no MC was detected (Table I). A similar proportion of each group were transplanted in remission (x2=0.30). Although the median grade of GvHU was different (0 against 2), this

* Royal Free Hospital Marrow Transplant Group: H. G. Prentice. H. A. Blacklock, M. Brenner. J .

t Bloomsbury Bone Marrow Transplant Group: A. H. Goldstone. J . D. M. Richards, J. C. Cawley. D. C. Patterson. G. Janosby. D. Skeggs, A. V. Hoflbrand.

Linch, C. C. Anderson, N. Godlee. J. Tobias, E. K. Huehns.

Host Haernopoietic Cells Detected bg Cylogenetir- Analysis of Bone Marrow 3 8 7

Table I. Clinical features of patients with and without mixed chimaerism

Detectable Donor cells host cells only

No. of patients Age range Sex

Diagnosis

Grafted in remission (%) Median GvHD grade

Royal free GvHD score Disease recurrence (%) Survivors (:(,)

10 29

18 (6-35) 21 (5-43) 4 male 20 male 6 female 9 female 5 ALL 1 6 A L L 3 ANLL 11 ANLI, 1 NHL 2NHL 1 CGL

6 (67'%)* 22 (76%)

0 2

0.2 5 0.34

6 (60%) 7 (24%)

4 (40%)t 7 (24%)8

Patient with chronic phasc CGL cxcluded. Two patients with recurrent disease.

:{ All patients disease free.

did not attain statistical significance by Wilcoxon's unpaired rank sum test, and the Royal Free GvHD score (Prentice P t al, 1984) for the groups was similar (0.25 against 0.34). Although a higher proportion of patients with MC have relapsed (60% against 24%; xL=4.30, P<O.O5), a similar proportion of each group are disease-free survivors (20% against 24%; x2 =0*07) .

Details of the cytogenetic studirs and the clinical outcome of patients with MC are given in Table 11. Seven of 3 1 patients (23%) examined within 6 weeks of RMT had MC. A mean of 9 "/o (range 3-1 5%) recipient metaphases were detected. In three patients (4, 6 and 7) recipient cells persisted beyond 6 weeks. Patient 7, who had CGL, displayed an increasing proportion of Philadelphia chromosome positive cells. Patient 4 hsd a single recipient cell detected on one of three cytogenetic examinations after 15% host metaphases had been found at day 14, and remains in remission at > 632 d. Patient 6, who was transplanted in relapse, subsequently displayed an increasing proportion of recipient metaphases. She relapsed 24 7 d after initial detection of MC. In four patients (1-4), recipient metaphaxes were undetectable at their last cytogenetic examination. Patients 2 and 4 remain alive and disease-free (>470 and > 632 d). Patients 1 and 5 died of infective complications without evidence of relapse. Patient 3 was transplanted in relapse arid had 1 3 % hyperdiploid recipient metaphases on day 22. Only donor metaphases were detected at 41 d but she relapsed on day 148. Cytogenetic analysis was not performed at relapse.

3 8 8 Helen Walker et a1

Table 11. Details of patients with mixed haemopoietic chimaerism

GvHD* Marrow analyses Present Status grade/ ~ ~ ~ ~ Relapse status

IJPN Age/sex Diagnosis at RMT pr’xis Day Host cells/total (day) (day)

Early MC 1 AL30 31/M

2 AL87 6/M

3 AL92 7/F

4 AL79 26/M

5 AL86 13iM

6 AL93 35/F

7 UC83 l l / F

Late MC 8 AL17 19/F

9 AL78 26/F

10 AL3 9/F

~

ALL

ALL

ALL

ANLL

ANLL

(51) ANLL

CGI,

ALL

Ar,r.

NHL

First I/a 20 remission 24

67 First O/b 16 remission 42 Relapse O/b 22

42 First O/b 14 remission 64

130 168

Relapse IV/c 14 3 1

Relapse O/b 1 5 75 98

118 Chronic Old 19 phase 50

113

Second i/a 14 remission 28

155 238

Second O/b 1 7 7 remission 337 First II/c 47 remission

0110 2/26 012 8 3/32 012 7

4I3O.t 0130 3/20 0120 1/40 0/40

01 7 1/40

513 3 14/20 28/30 27/28

3 1 ~ 1124s

014

54/74$

0113 8/30

18/22 1612 7

8/8 15/20

No Alive

Yes Dead (148) (178)

No Alive

(>470)

(>632)

No Dead

Yes Alive (262) ( > 3 9 9 )

Yes Alive (115) (>715)

Yes Dead (191) (311)

Yes Dead (337) (378) Yes Dead ( 7 8 ) (110)

* GvHD grade (Thomas et al, 1975). GvHU prophylaxis: a: in vitro opsonization with OKT3 +post-graft methotrexate (Prentice et al, 1982): b: in vitro lympholysis with MBG6 + RFT8 +rabbit complement (Prentice et al, 1984); c: post-graft methotrexate; d: post-graft cyclosporin A.

t Hyperdiploid clone. $ Majority Philadelphia chromosome positive. 3 All Philadelphia chromosome positive.

Host Hueinopoietit Cells Detected by Cytogenetic Ariulysis of Bone Marrow 3 89

I n three patients (8-10). MC (mean 54% recipient metaphases) was first detected morc than 6 weeks after BMT. Patient 8 had been examined on two previous occasions without detection of recipient cells. The other two patients had not had previous cytogenetic examinations. The mean interval betwccn detection of recipient cells and diagnosis of relapse was 76 d (range 3 1-160 d).

Seven of the 37 patients studied relapsed without prior detection of MC. Negative cytogenetic examination preceeded diagnosis of relapse in these patients by a mean of 301 d (range 106-804). The patient conditioned with melphalan alone had no MC on four cytogenetic analyses after BMT, but reIapsed on day 11 56, 804 d after his last cytogenetic analysis.

DTSCrSSTON

In this study, 2 304 (7/3 1) of patients examined in the first 6 weeks after allogeneic BMT were found to have MC. Three have relapsed; two are disease-free at > 470 d and > 6 32 d. MC early after BMT therefore need not necessarily predict relapse. In contrast, all three patients in whom recipient cells were first identified morc than 6 weeks after BMT subsequently relapsed. The single patient (4) with MC after 6 weeks who did not reIapse had only 2.5% mosaicisrn and these recipient cells subsequently disappeared. The presencc of greater proportions o f MC later than 6 weeks after BMT appears to be strongly associated with subsequent relapse.

Santos et nl (1972. 1974) found persistent recipicnt cells up to 6 weeks after Cy conditioning, which is now considered insufficient ablative therapy (Thomas et aJ, 19 76: Souhanii et al , 1982; Smith Pt al. 1983). However, our finding of persistent recipient rnetaphases in 23% of patients within 6 weeks of BMT conditioned with Cy and TRI is higher than expected and has not been reported prcviously. Lawler ot nl(1984) did not find recipient metaphases within 9 weeks of BMT but this may have been due to differences in conditioning regimens or cytogenetic techniques. Santos et ul (1 9 72) interpreted their findings as providing strong evidence that recipient bone marrow cells were not completely eliminated by their conditioning regimen but by clinical or subclinical graft-versus-host disease (GvHD). This suggestion is supported by recent observations of differences in recurrence rates in patients receiving allogeneic and syngeneic BMT for AML (Gale & Champlin, 1984). However, we found no significant difference in the severity of GvHD seen in patients with or without MC. In addition there was no correlation between the occurrence of GvHL) and subsequent persistence or disappearance of recipient cells in patients with MC, although the number of patients studied may be too small to dctect any such relationship. The nature of the persisting popuIations of recipient fells has not been determined and their disappearance in some patients may be a result of incapacity for self-renewal rather than GvHD.

Further evidence of the inadequacy of Cy and TBI conditioning has been provided by Branch et aJ (1982). Four cases of MC between 238 and 553 d post-BMT were identified by red cell phenotyping (range 2-99% recipient cells) and cytogenetic studies of peripheral blood lymphocytes (up to 80% recipient origin). These patients were followed for a mean period o f 1 20 d (range 2 3-2 70 d) and none relapsed. All four patients in our study with more than 3% mosaicism more than 6 weeks after BMT have relapsed between 78 and 33 7 d. We have not

390 Helen Walker et a1

seen patients transplanted for acute leukaemia with MC at an equivalent period to those of Branch et a1 (I 982). The different experience of recurrent disease in the two studies may be due to the small number of patients studied, differences in conditioning regimens or remission status at transplant. However, it is possible that detection of recipient metaphases in bone marrow may be more indicative of subsequent recurrence than are red cells of host phenotype or lymphocytes of host origin in the peripheral blood.

Patients 9 and 10, in whom MC was first detected more than 6 weeks after BMT, may have shown similar repopulation with host cells to that seen in patient 6, had earlier examinations been performed. Patient 8 had no detectable recipient cells on two earlier examination but these only excluded mosaicism over 2 1 % with 9 5% confidence (Hook, 19 77) . This case emphasizes the limitations of cytogenetic analysis in detecting mosaicism of small proportions, and the importance of examining large numbers of metaphases to do so.

Comparison of outcome in patients with MC with those in whom no recipient metaphases were detected is difficult because of a high early mortality in the latter group. These patients died of GvHD and associated infections before more effective prophylaxis was developed (Prentice et a1, 1984). The seven patients in this group who relapsed without prjor detection of MC, had a long interval (mean 301 d) between their last cytogenetic examination and diagnosis of recurrence. In contrast, the group in whom MC was detected prior to recurrence had a mean interval of 88 d. More frequent cytogenetic assessment of the group who relapsed without MC might have permitted detection of recipient metaphases prior to recurrence.

In addition to time of detection in relation to BMT, three other factors appear to be of prognostic value in a patient with MC. Firstly, the proportion of host cells in the bone marrow may determine the probability both of their persistence and their containing a proportion of clonogenic leukaemic cells. A mean of 9% recipient metaphases were detected in the four patients who did not subsequently relapse while the six patients who did relapse had a mean of 32% recipient cells at first detection. A second factor may be the presence of abnormalities in the recipient karyotype detected. Finally, an increasing proportion of recipient metaphases appears to be a poor prognostic sign. Three patients with early MC relapsed: one had a low proportion of abnormal recipient cells and the others had increasing numbers of recipient metaphases.

This study reveals that MC is not uncommon soon after BMT but need not necessarily indicate a poor prognosis. The conditioning regimen and remission status at BMT may affect the incidence of MC. Four factors have been identified which may help determine the risk of relapse in patients with MC. Cytogenic examination of bone marrow after BMT thus appears to be of value in identifying patients with a high risk of recurrent disease. Frequent cytogenetic assessment of bone marrow may also provide the earliest indication of recurrence in these patients. This may enable treatment of a low tumour load when effective salvage therapy is available for patients who relapse after BMT. In the meantime a more ablative conditioning regimen may have to be sought.

Host Haernopoietic Cells Detected by Cytogenetic Analysis of Bone Marrow

REFERENCES

391

BRANCH, D.R., GALLACHER, R/I.T., FORMAN. S.J.. WINKI~ER, K.J., PETZ, L.D. & BLLME. K.G. (1982) Endogenous stem cell repopulation resulting in mixed haemopoietic chimerism following total body irradiation and marrow transplantation for acute leiikaemia. Tran.splantation, 34, 226- 228.

GALE, R.P. S- CHAMPLIN. K.E. (1984) How does bone marrow transplantation cure leukaemia? Lancet, ii, 28--30.

HOOK, E.B. (1977) Exclusion of chromosome mosaicism: tables of 90%. 95% and 99% confi- dence limits and comments on use. American journal of Hurnnn Genetics, 29, 94-97.

ISCN (1 978) An international system for human cytogenetic nomenclature. Birth Defects: Origi- rial Article Series, 14, No. 8. The National Foundation. New York.

LAWLER, S.D., BAKEK, M.C., HAKRIS. H. b; MORGEN- STERN. G.R. (1984) Cytogenetic studies on reci- pients of allogeneic bone marrow using sex chromosomes as markers of cellular origin. British lournal of Huernutology. 56, 4 3 1-443.

PRENTICE. H.G.. BLACKLOCK, H.A., JANOSSY. G.,

HOFFBRAND. A.V. (1982) Usc of anti-T-cell monoclonal antibody OKr3 to prevent acute graft-versus-host disease in allogeneic bone marrow transplantation for acute leukaemia. Lancet. i. 700-703.

PRENTICE. H.G., BLACKLOCK. H.A., JANOSSY. G., GILMORE. M.J.M.L.. PRICE-]ONES. I,., TIDMAN, N., 'I'REJI)OSIEWICZ, L.K., SKEGGS, D.B.L., PANJW-AN, D., BALL. S.. CRAPHAKOS. S.. PATTERSON. J , , IVOKY. K. & HOFFBRAND. A.V. (1984) Depletion of T- lyniphocytes in donor marrow prevents signifi- cant graft-versus-host disease in matched allo- geneic leukaemic marrow transplant recipients. Lancet. i , 472-476.

SANIOS. G.W.. SENSENBRIJNNER, L.L., BIJRKE. P.J.,

BRADSTOCK. K.F.. SKEGGS, D., GOT.T)STEIN, G. &

MULLIXS, G.hf., ANDERSON, k'.N., TIJTSCHKA, P.J., BRAIKE. H.G., D a m . T.E., HUMPHREY, R.I,.. ABELOFF. M.D. BIAS, W.B., BOKGAONKAR, D.S. & SLAVIN. R .E. ( 1 9 74) Allogeneic marrow grafts in

man using cyclophosphamide. Transplantation Proceedings, 6, 345-348.

SANTOS. G.W., SENSENBKUNNER, T L . , BURKE, P.J., MULLINS, G.M.. BIAS. W.B.. TUTSCHKA, P.J. & SLAVIN, K.E. (1972) The use of cyclophospha- mide for clinical marrow transplantation. Trans- plantation Proceedings, 4, 5 59-564.

SANTOS, G.W. & TUTSCHKA. P.J. (1972) Allogeneic marrow transplantalion in a rat model of clini- cal aplasia. Federation Proceedings. 31, 233.

J.T.. (1983) Autologous bone marrow rescue is unnecessary after very-high-dose cyclophos- phamide. Lancet. i. 76-77.

SOUHAMI. R.L., HARPER, P.G., LINCH, U.C., TRASK, C.. GOLDSTONE, A.H., TOBIAS. J.S., SPIKO, S.G., GEDDES. D.M. S- RICHARDS, J.D.M. (1982) High dose cyclophosphamide with autologous mar- row transplantation as initial therapy of small cell carcinoma of bronchus. Crincer Cherno- therupy und Phurmacology. 8. 3 1-34.

SPARKES, M.C., CRIST, M.L.. SPAKKES. K.S., GALE. R.P.. FEIG, S.A., UCLA TRAMPLANTATION GROUP ( 1 9 771 Gene markers in human marrow trans- plantation. Vox Sanguinis, 33, 202-205.

SUMMER. A.T., EVANS, H.G. S- BUCKLANU, R.A. (1971) New techniques for distinguishing between human chroniosomes. ,Vaturr: New

THOMAS, E.D., STORB, R.. CLIVT. K.A.. FBFER. A,. JOH~SSON, F.I,., NEIMAN, P.E., LERNER. K.G., GLIJCKSBEKG, H. & BUCKNER. C.D. (1975) Bone marrow transplantation. Nrw Enylrmd ]uurnal of Medicinr. 292, 832-8431895-902.

THOMAS, E.U.. STORR, R.. GIBLETT, E.K.. I,oN(;I'RE, R.. WEIDEN, P.L., FFFER, A., WITHEKSPO~N. R., WFT, R.A. b; BI:CKNER, C.D. ( 1 976) Recovery from aplastic anaemia following attempted marrow transplantation. Experimental Harmatology, 4,

'rJI0, J.H. & WHahc, J. (1962) Chromosome prep- aration of bone marrow cells without prior in vitro culture or in vivo colchicine administra- tion. Stain Technology, 37, 17-20.

ShlITH. I.E., EVANS, B.D., HARLAND, S.J. & MILLAR.

B i ~ l ~ g . ~ j , 232, 31-32.

9 7-1 0 2 .