Central nervous system involvement at presentation in acute granulocytic leukemia

Central Nervous System Involvement at Presentation in Acute Granulocytic Leukemia A Prospective Cytocentrifuge Study

RICHARD J. MEYER, M.D. PLACID0 PAUL0 C. FERREIRA. M.D. JANET CU’ITNER. M.D. MICHAEL L. GREENBERG, M.D. JUDITH GOLDBERG, Sc.D. JAMES F. HOLLAND, M.D.

NewYork, New York

We have undertaken a perspective study of the prevelance of the central nervous disease in acute granulocytic leukemia (AGLj Thirty-nine newly diagnosed patients with AGL underwent cytocentrifuge examination of cerebral spinal fluid. Seven of the 39 patients had blast cells in their cerebral spinal fluid. All seven of these patients had acute myelomonocytic leukemia (AMMLj No patients with other variants of AGL demonstrated blast cells in their cerebral spinal fluid. Other high risk factors associated with meningeal infiltration were elevated serum lysoyme levels, high peripheral white blood cell count, low age, splemomegaly and the presence of infiltration in other organs. The admission rates for patients with meningeal leukemia were lower and the survival iime was shorter than in both the 32 noninvolved patients and the noninvolved paGents with AMML. We believe that a lumbar puncture is indicated in all patients with newly diagnosed AMML. Central nervous system involvement in childhood acute lymphocytic leukemia (ALL) was not a clinical problem until the advent of effective chemotherapy. [1,2] With increasing length of survival, relapse with central nervous system involvement rose to 56 to 83 per cent [3-81. With prolongation of survival, an increasing incidence of leukemic infiltration of the meninges in patients in the terminal phase of chronic

From the Division of Hematology, Department of Medicine, Department of Biostatistics and Department of Neoplastic Diseases. Mount Sinai School of Medicine, New York, New York. This study was.supported in part by Grants CA 04457 and CA 10515 from the National Cancer Institute. It was presented in part at the 19th Annual Meeting of the American Society of Clinical Oncology, Washington, D.C.. 1978. Requests for reprints should be addressed to Dr. Janet Cuttner. Mount Sinai School of Medicine, Fifth Avenue and 100th Street. New York, New York 10029. Manuscript accepted October 19,1979.

myelogenous leukemia (CML) has been reported [g-16]. In childhood acute LTanulocytic leukemia (ACL) Fore effective therapy has resulted in increasing numbers of relapses in the central nervous system [17,18]. Although there appears to be an increasing frequency of meningeal leukemia in adults with ACL [19-231, the true incidence is undetermined. With the knowledge that central nervous system relapse results from the growth of microscopic leukemic foci present at the time of initial systemic treatment. we have undertaken a prospective study of the prevalence of leukemic meningitis at presentation in AGL. MATERIALS AND METHODS Thirty-nine consecutive patients newly diagnosed as having acute myelocytic (AML), promyelocytic (APL). erythro (ERL) or myelomonocytic (AMML) leukemia according to the criteria of Hayhoe and Cawley 1241and Hillestad 1251 underwent a lumbar puncture by one of us prior to the institution of identical systemic therapy with daunorubicin and cytosine arabinoside. We collected 1.5 ml of spinal fluid into 0.5 ml of fetal calf serum. The fluid was immediately processed. Aliquots of 1.0 ml were placed in Shandon Elliot cytocentrifuge

May 1660

The American Journal of Medicine

Volume 66

661

CENTRAL

TABLE I

NERVOUS

SYSTEM INVOLVEMENT

IN ACUTE GRANULOCYTIC

LEIJKEMIA--LlEYER

ET AI..

The Clinical Parameters of Seven Patlents with Blast Cells Demonstrable on lnltlal Lumbar Puncture Temperature

Patient

Cell Type

1 2

AMML AWL AMML AMML AMML AMML AMML

3 4 5 6 7

Sex and Age (yr) F, M, F, M, M. M, F.

al Presentation (F”)

SplenwneOaly

103 98 100.4 103.8 98 98 98

+ + + 0 + + +

30 64 54 64 59 58 17

Organ Infiltration +G

+G, S +G, S 0 0

Neurokgk

Msuroiogk

Symptoms

Sions

H H 0 OMS 0

F 0 0 F 0

0 +G

0

0

Dip

NYS

Status AHer Systemic ThefepY CR -237

days

Alive-no CR CR-340 Died-no Died-no Died-no Died-no

+ CR CR CR CR

NOTE: S = skin infiltrates, G = gingival hyperplasia, H = headache, F = fundal hemorrhages and exudates, OMS = organic mental syndrome, Dip = diplopia, NYS = nystegmus, CR = complete remission.

chambers and centrifuged at 100 rpm for 4 minutes onto glass slides previously coated with 1 per cent bovine serum albumen. The resultant slides were air dried and stained with Jenner-Giemsa. They were interpreted independently by three members of the group. Except as noted, patients with blast cells in the cerebrospinal fluid were treated with intrathecal cytosine arabinoside, 30 mg/m2 body sufface area (maximum dose 50 mg) [26,27]. each week for five weekly doses or until clearing of blast cells from the spinal fluid if residual blast cells were still present after five weeks. In following the 22 patients who entered complete remission 11 underwent repeat lumbar puncture with cytology while in sustained remission of from 90 to 512 days. Twelve of the 22 patients underwent repeat lumbar puncture plus cytocentrifye study at the time of initial systemic relapse from 100to 540 days after initial lumbar puncture. There were no clinical sequelae of lumbar puncture. either in the initial tap or in the group undergoing intrathecal therapy. The patients with and without central nervous system involvement arc compared, using t tests and chi-square tests to assess differences. Since there are small numbers of patients and multiple t tests were performed, associated P values are used as additional differences.

summary

information

for suggestions

of

RESULTS

Seven of the 39 patients had demonstrable blast cells in the cerebrospinal fluid (Tables I and II]. Sex did not distinguish the two groups. There were four male and

TABLE II

The Laboratory Parameters of Seven Patients with Blast Cells Demonstrable on lnltlal Lumbar Puncture

PatW

Gel! lope

1 2 3 4 5 6 7

AMML AMML AMML AMML AMML AMML AMML

692

three female patients with central nervous system involvement [group l] compared to 17 male and 15 female patients without central nervous system involvement (group 21. The mean white blood cell count in the group with central nervous system involvement [group 1) was higher than that in the group without such involvement (group 2) -98.3 X lo”/pl (range 10.7 to 175 X 103/~1) versus 37.4 X 103/~J (range 1.2 to 458 X 103/~1] (t 37 = 1.8, 0.10 > P > 0.05). Mean platelet counts in the two groups did not differ -83.6 X 103/pl (range 30 to 175 X 103/pl] in group 1 ai compared to 84.6 X 103/pl (range 10 to 218 X 103/pl) in group 2. A highly significant difference in the serum lysozyme level existed between the two groups (t 34 = 3.8, p < O.OOl]. Mean serum lysozyme in group 1 was 72.9 pg/pl (range 42 to 110 pg/pl] compared to 30.3 I.cg/pl (range 7 to 120 rg/pl] in group 2. The levels of both cerebrospinal fluid protein or sugar were similar in both groups of patients. Splenomegaly was noted in six of seven patients in group 1 and in 12 of 32 patients in group 2 (xl2 = 5.4, 0.025 > P > 0.01). Extravascular infiltrations, as suggested by gingival hyperplasia and/or infiltrative skin lesions, were found in four of seven patients in group 1 and in two of 32 patients in group 2 (xl2 = 11.4, P < 0.005]. The mean age of the seven patients (group l] was 49.4 years and of the 32 patients (group 2) 56.8 years (t 37 = 1.03, 0.40 > P > 0.20). All patients with blast cells in the cerebrospinal fluid

May 1980

Sex and Age (yr) ‘F, M, F, M, M, M. F.

30

64 54 64 59 58 17

White Blood CellCoulli (lo-s/@)

Pl4ltOlOt (lo-‘/@)

Serum LrsOqm (rdml)

137 175 175 10.7 59.5 77 54

169 24 30 175 63 70 54

94 98 110 60 42 52 54

The American Journal of Medicine

Volume 68

Cefeluoephal hid Prolrh (mollW (mgllgg mU 24 47 46 66 47 45 11

74 65 97 a0 71 107 57

White Bbd cell coant mU

(prr CtU 13 4 1 70 1 1 1

CENTRAL

NERVOUS

SYSTEM

INVOLVEMENT

had AMML (P < 0.005). No patient with AML or APL had blast cells in the cerebrospitial fluid. Presence or absence of an elevated temperature did not separate the two groups. The only patient with cranial nerve palsy had blast cells in the cerebrospinal fluid. One other patient in group 1 presented with confusion. Headache was a common symptom in both groups and did not separate the two groups. Although papilledema was present in only one patient, a patient in group 1,fundal hemorrhages and exudates were so common as not to differentiate the groups. Two of seven patients with cerebrospinal fluid involvement entered complete remission. One patient continued in complete remission for 340+ days. Twenty of 32 patients without involvement entered complete remission for from 66 to 602+ days. No patient in this study had a relapse in the central nervous system (xl2 = 2.7, 0.25 > P > 0.10). No patient exhibited blast cells in their spinal fluid upon repeat lumbar puncture while in remission or at the time of systemic relapse. No patient had a relapse in the central nervous system. No systematic effort was made to obtain chromosome analysis on the effected group. However, in the four patients in whom it was performed the Philadelphia chromosome was absent.

COMMENTS

In a review of central nervous system involvement in leukemia in 1935 Schwab and Weiss [l]stated that involvement of the central nervous system in leukemia was an unusual complication mentioned as rare or not described in standard textbooks. Evans [2] found only a 4 per cent incidence of meningeal leukemia in childhood ALL in 1948.With prolongation of survival in ALL, the incidence approached 4 per cent per month [4] with up to 83 per cent of the children having a relapse in the central nervous system [3]. The introduction of “prophylactic” therapy with intrathecal methotrexate with or without central nervous system irradiation significantly reduced this incidence, opening the possibility of “cure” in ALL [28,29]. It must be emphasized that early meningeal treatment is not prophylaxis but the destruction of blast cells present in numbers below our previous ability to detect by the classic method of analysis, i.e., in the hemocytometer chamber. Evans reported that in 10 of 116 cases of meningeal leukemia cell counts were “normal.” Several cases of autopsy-proved central nervous system infiltration with no cerebrospinal fluid pleocytosis have been reported in AGL (301. Use of more refined techniques such as millipore filtration [31] and use of the cytocentrifuge [32,33] have greatly increased diagnostic sensitivity. In childhood ALL Evans et al. [32] have shown that 30 per cent of the specimens with normal cell counts contain leukemic cells when slides prepared on the cytocentrifuge were examined. These results are similar to the findings of May

IN ACUTE

GRANULOCYTIC

LEUKEMIA-MEYER

ET AL.

Drewinko et al. [33]. We confirm these findings in AGL. One of our patients with “normal” hemocytometer cell counts, but with blast cells, had clinically evident central nervous system leukemia, in one clinical central nervous system leukemia and pleocytosis developed when not treated, and in one other increasing cerebrospinal fluid pleocytosis developed. In two other cases increasing cerebrospinal fluid protein developed, and in one nuchal rigidity and papilledema were present with only 2 blast cells/ul. In no patient in group 2 did central nervous system leukemia or abnormalities in the cerebrospinal fluid develop, with follow-ups up to 540 days. Demonstration of blast cells on cytologic preparations is diagnostic of meningeal leukemia [3]. Our total prevalence of 20 per cent for leukemic meningeal infiltration in AGL approximates that seen at autopsy [13] and in some clinical series [19-,211,but it is far higher than that reported by Wolk et al. [13]. The findings of increasing leukemic meningitis with high peripheral white blood cell count and organ infiltration parallels the findings of Pavlovsky et al. [34] in children and adults, and of West [6] in children. Patients with high serum lysozyme levels had a significantly higher risk of meningeal infiltration (p <0.005). A correlation between development of asymptomatic central nervous system leukemia late in remission and a high serum lysozyme level has recently been reported in an abstract by Peterson et al. [21]. Of special importance is the definition of a group with great risk for the presence of central nervous system disease, i.e., those patients morphologically characterized as having AMML. Cells present in AMML may have a greater tendency to cross the blood brain barrier as a stochastic process associated with greater numbers in the blood, or by virtue of metabolic or other functional differences from other types of leukemia. Splenomegaly and skin infiltrations may be other manifestations of the same factor(s) responsible for infiltration into the cerebrospinal fluid. Patients with meningeal leukemia at presentation appear to have less chance of entering complete remission. We were unable to identify other contributing factors. With the identification of this subgroup at greater risk we suggest that a lumbar puncture with slides prepared on the cytocentrifuge be performed on all such patients. Prompt therapy has resulted in prolonged eradication of nervous system disease in our patients as in other series [20,21]. The role for “prophylactic” treatment in these patients is less clear. Patients have been described with ALL and in the blastic phase of CML with clinically apparent leukemia limited to the central nervous system in whom leukemia subsequently developed in the blood and marrow [IO]. Initial sanctuary treatment has not been shown to prolong survival in children with AML [35]. The role of central nervous system therapy at diagnosis in the group of patients at high risk, i.e., with AMML and/or elevated serum lysozyme levels should be subjected to prospective study. 1960

The American Journal of Medicine

Volume 66

693

CENTRAL NERVOUS SYSTEM INVOLVEMENT IN ACUTE CRANULOCYTIC

LEUKEMIA-MEYER

ET AL.

REFERENCES 1. Schwab RS. Weiss S: The neurologic aspects of leukemia. Am J Med Sci 1935; 189:766. 2. Evans AE: Central nervous system involvement in children with acute leukemia. A study of 921 patients. Cancer 1964: 17:256. 3. Pinkel D: Five-year followup of “total therapy” of childhood lymphocytic leukemia. JAMA 1971: 216:648. 4. Evans AE, Gilbert ES, Zaudstra R: The increasing incidence of central nervous svstem leukemia in children. Cancer 1970: 26:404. ’ 5. Haghbin M, Zeulzcr WW: A long-term study of cerebospinal leukemia. J Pediatr 1965; 6723. 6. West RI. Pole IG. Hardistv RM. Pike M: Factors in the pathogencsis of central nervous’system leukemia. Br Med J 1972.311. 7. Nies BA. Thomas LB, Freireich EJ: Meningeal leukemia-a followup study. Cancer 1965: 18546. 8. Hyman CB, Boyle JM, Brubaker CA, Williams K. Hammond D: Central nervous system involvement by leukemia in children in relationship to systemic leukemia and description of clinical and laboratory manifestations. Blood 1965: 25:l. 9. Morrow GW. Jr, Place GL, Schoebel CF, Bcnnet WA: Terminal phase of chronic myclogenous leukemia. Cancer 1965; I&369. 10. Atkinson K. Kay HEM, Lawler SD, Wells DG, McElwain TJ: Meningeal leukemia after blastic transformation of chronic myeloid leukemia. Cancer 1975; 35:529. 11. Schwartz IH. Canellos G. Younn RC. DeVita VT: Meninneal leukemia in the blastic phase of chronic granulocytic Teukcmia. Am J Med 1975: 59:819. 12. Kwaan HC. Pierre RV. Long DL: Meningeal involvement as first manifestation of acute mycloblastic transformation in chronic granulocytic leukemia. Blood 1968; 33:348. 13. Wolk RW, Masse SR, Conklin R, Freireich EJ: The incidence of central nervous system leukemia in adults with acute leukemia. Cancer 1974:33:863. 14. Bornstein RS. Nesbit M, Kennedy BJ: Chronic myclogenous leukemia presenting in blastic crisis, Cancer 1972; 30: 939. 15. Mastrangclo R, Zuclzer WW. Thompson R: The significance of the Ph chromosome in acute myeloblastic leukemia. Serial cytogenetic studies in a critical case. Pediatrics 1967: 40634. 16. Lui PI, Ishimaru T. McGregor DH: Autopsy study of blast crisis in patients with chronic granulocytic leukemia. Hiroshima and Nagasaki 1949-1969. Cancer 1974: 33:1062. 17. Fleming I. Simone 1, lackson R. lohnson W. Walters T, Mason C: Splenectomy~ &Id chemotherapy in acute myelocytic leukemia of childhood. Cancer 1974: 33:427. 18. Walters T, Aur R, Hcrnandez K, Vietti T, Pinkel D: SixAzauridine in combination chemotherapy of childhood acute myelocytic leukemia. Cancer 1972: 29:1057. 19. Dawson DM, Rosenthal DS. Moloney WC: ‘Neurological complications of acute lcukcmia in adults. Changing role.

694

May 1980

The American

Journal of Medicine

20.

21.

22.

23.

24. 25. 26. 27. 20. 29. 30. 31. 32. 33. 34. 35.

Volume 68,

Ann Intern Med 1973; 79:54I. Wiernik P, Schimpff SC. Schiffer CA, et al.: Randomized clinical comparison of daunorubicin alone with a combination of daunorubicin, cytosine arabinoside. 6-thioguanine and pvramcthamine for the treatment of acute nonlymphocytic leukemia. Cancer Treat Rep 1976: 60:41. Peterson BA. Bloomficld CD: Asvmotomatic central nervous system leukemia in adults with’acutc non-lymphocytic leukemia in extended remission labstract), American So’ ciety of Clinical Oncology 1977: 341, Mever RI. Ferreira PPC. Cuttner 1. Greenbern ML. Holland JF: A prospective cytocentrifuge study of c&tral nervous system leukemia at presentation in acute granulocytic leukemia [AGL]. Evidence for higher incidence in acute myclomonocytic leukemia (AMML). Proceedings of the American Society of Clinical Oncology 1978: 19: 350. Ferreira PPC, Cuttner J. Meyer RJ, Holland JF: Increased central nervous system involvement in acute myelomonocytic leukemia-Cytocentrifuge studies. Proceedings of the 19th Annual Meeting of the American Society of Hematology, Boston 1976: 129. Hayhoc FGJ, Cawley JC: Acute leukemia. Cellular morphology, cytochemistry and fine structure. Clin Haematol 1972: 1: 49. Hillestad LK: Acute promyclocytic leukemia. Acta Med Stand 1957; 2: 641. Wang JJ, Pratt CB: Intrathecal arabinosyl cytosine in mehingeal leukemia. Cancer 1970: 25:531. Band PR, Holland JF, Bernard J, Weil M, Walker M. Rall D: Treatment of central nervous system leukemia with intrathecal cytosine arabinoside. Cancer 1973; 32:744. Simone J. Aur R, Hustu H. Verzosa M: Acute lymphocytic leukemia in children. Cancer 1975; 36:770. Hustu H, Aur R, Verzosa MS, Simone J. Pinkel D: Preventions of central nervous system leukemia by irradiation. Cancer 1973; 32:585. Rawbone RG, Shaw MT, Bagshawe KD: Intracranial leukemia and the cerebrospinal fluid. Br Med ,I 1972:20: 444. DelVecchio PR. Dewitt SH. Borelli JI, Ward JB. Wood, TA, Malgrcns RA: Application of millipore filtrations technique to cytologic material. J Nat Cancer lnst 1959: 22:427. Evans DIK. O’Rourke C. Morris-Jones P: The ccrebrospinal fluid in acute leukemia of childhood. Studies with cytocentrifuge. J Clin Pathol 1974; 27:226. Drewinko B. Sullivan MP. Martin T: Use of the cytocentrifuge in the diagnosis of meningeal leukemia. Cancer 1973: 31: 1331. Pavlovsky S, Eppinger-Helft M, Muriel FS: Factors that influence the appearance of central nervous system lcukemia. Blood 1973; 42:935. Dahl G. Simone J, Hustu H, Mason C: Preventive CNS irradiation in acute myelogenous leukemia. Proceedings of the 19th Annual Meeting of the American Society of Hematology, Boston. 1976; 32.