PROGNOSTIC FACTOR ANALYSIS IN ADULT ACUTE LEUKEMIA DURING THE PERIOD OF 1973 TO 1983

Transcription

1 Keio Journal of Medicine 34: 17-23, 1985 PROGNOSTIC FACTOR ANALYSIS IN ADULT ACUTE LEUKEMIA DURING THE PERIOD OF 1973 TO 1983 HAYATO MIYACHI, YUZURU TAKEMURA, KEISUKE TOYAMA and TOSHIAKI HIGASHI*, Department of Internal Medicine *1 Department of Preventive Medicine and Public Health, School of Medicine, Keio University, Tokyo, Japan (Received for publication December 6, 1984) ABSTRACT The pretreatment characteristics of 85 adults with acute leukemia who were treated at Keio University Hospital between 1973 and March 1983 have been evaluated to assess their value as prognostic indicators. Three factors were found to be associated with probability of obtaining a complete remission. We con firmed characteristics previously known to provide prognostic information such as morphology, temperature status and hemoglobin level using the chi-square test. A linear correlation was found between serum lactic dehydrogenase level and blood cell counts, and between platelet counts and nucleated cell counts in the bone marrow. Key words: Prognostic factor, Adult acute leukemia INTRODUCTION The ultimate aim in the treatment of acute leukemia is total cure of the patient; the attainment of complete remission is the initial step of this goal. The survival curve for failing patients is almost identical to that of patients receiving supportive care only., The median survival period after relapse is generally less than six months,2 so that once hematologic remission is achieved, the next therapeutic objective is to prevent recurrence of leukemia. In treatment, therefore, it is important to evaluate case charac teristics related to prognoses. Recently the proportion of patients having a five-year survivorship has greatly increased, and a number of studies have been published on pretreatment factors helpful in predicting either the likelihood of achieving complete remission or the duration of remission.3 17

2 18 Hayato Miyachi et al. We carried out this study in order to investigate the relationship between prognoses and case characteristics of 85 patients with acute leukemia prior to treatment. MATERIALS AND METHODS Between 1973 and March 1983, 85 acute leukemia patients with no previous his tory of intensive treatment underwent remission induction treatment at Keio University Hospital. The patients varied in age from 14 to 80 years (median: 43). All of them were subclassified using the French-American-British (FAB) criteria. Patients with chronic myelocytic leukemia in blastic transformation, erythroleukemia (M6), and smouldering leukemia were excluded from analysis. For primary remission induction treatment, patients with nonlymphocytic acute leukemia underwent relative homogenous chemotherapy with a combination of dauno rubicin (DNR), cytosine arabinoside (CA), 6-mercaptopurine (6-MP), and predni solone (PSL); vincristine (VCR) was added for patients with lymphocytic leukemia. Complete remission was diagnosed upon the attainment Z 80 points on the Kimura's remission criteria's together with the absence of blasts in the peripheral blood. The patient characteristics chosen for analysis were recorded at the time of diag nosis, not upon the initiation of chemotherapy. These included morphology, age, sex, body temperature, liver size, hemoglobin level, white blood cell (wbc) count, platelet count, neutrophil alkaline phosphatase (NAP) score and lactic dehydrogenase (LDH) level. To assess the possible association of dichotomous variables with remission inci dence, the proportion of patients who achieved complete remission was computed for the two subgroups defined by each variables. To compare these two proportions, a chi-square test was used. To define clearly the interrelationship a correlation-coefficient was computed. between the variables, RESULTS Sixty-seven (67%) of the 85 patients achieved complete remission during the period covered by this study. Remission rates for the morphological diagnoses were as follows: Ll, 12/12 (100%); L2, 5/8 (63%); Ml, 4/10 (40%); M2,25/34 (74%); M3,7/14 (50%); M4,1/3 (33%); M5,3/4 (75%) (Table 1). A significantly higher (p<0.05) response rate was found for L1. Remission duration and survival did not significantly differ from each other. Representative data for all patients are shown in Table 2. The proportions of patients who achieved complete remission for the two subgroups of each variables are shown in Table 3. The absence of fever was found to be the significant predictive factor (p<0.05) for the achievement of complete re mission, and the hemoglobin level was thought to be associated (p<0.1) with the complete remission rate. Initial LDH levels were correlated with wbc counts and platelet counts with nucleated cell counts in bone marrow (Table 4).

3 Prognostic Factor in Adult Acute Leukemia 19 Table 1 FAB subtype and prognoses Table 2 Representative data in all patients

4 20 Hayato Miyachi et al. Table 3 Response rate by pretreatment characteristics Table 4 Correlation-coefficients between pretreatment characteristics (N=23) * Significant at 0.05 level.

5 Prognostic Factor in Adult Acute Leukemia 21 DISCUSSION Such divergent factors as morphology, age, the presence of a documented ante cedent hematologic disorder, body temperature status, liver size, hemoglobin level, platelet count, the presence of Auer rods, chromosome pattern, LDH level, blood urea nitrogen, and creatinine level have been identified as major prognostic variables.5-1a We confirmed some of these. A high remission rate was observed among patients with LI, no fever, and a normal hemoglobin level. Leimert et al.11 have reported that LI showed a higher remission rate in adult acute leukemia. Similar trends have been noted in juvenile patients. M2, M4, or the presence of Auer rods have previously been shown to warrant prognoses for myeloid groups.12, Y3 Although the prognostic significance of Auer rods was not comfirmed for our group of patients, a higher response rate for LI was comfirmed. An intriguing ob servation is that although the complete remission rate for L1 was higher (100%) than M3 (50%), the median remission durations were similar. M3 has been reported to have a longer remission duration and a long survival period once complete remission is attained.13,14 While several factors were shown to influence strongly the likelihood of a patient's achieving complete remission, once the disease was in remission, these factors had little or no effect on remission duration or the likelihood of a five-year survival period.15 This effect is operative in such factors as body temperature status, hemoglobin level, and the presence of an antecedent hematologic disorder. Age has long been thought to affect prognoses adversely,5-10 But Gale et al. re ported a high (84%) remission rate with intensive induction chemotherapy, regardless of a patient's age, suggesting that even elderly patients could tolerate the period of hypoplasia. We also found no influence of age on complete remission rates. Intensive combination chemotherapy and improved supportive measures seem to alter the prog nostic significance of age. LDH level is more frequently elevated in patients with a high wbc count and extramedullary disease." LDH activity falls following effective treatment and closely parallels clinical status.17 In our group of patients, LDH levels prior to treatment were positively correlated with wbc counts. LDH level is strongly predictive of remis sion duration but not of the complete remission rate. Both platelet counts9 and fibrinogen levels5 were found to show this same tendency. A low platelet count has been reported by Passe et al.9 to indicate a longer remission duration. Low platelet counts are not, however, associated with specific FAB subgroups. It is conceivable that these may be specific leukemic phenotypes with good prognosis, but not recognized by conventional morphological criteria. We found platelet counts to be positively corre lated with nucleated cell counts in bone marrow. A high platelet count in spite of the leukemic replacement of bone marrow may indicate involvement of the megakaryocyte lineage in the leukemic process.18.

6 22 Hayato Miyachi et al. Several previous studies have identified prognostic factors, but all points remain controversial. Studies of the age factor have produced conflicting results. In contrast, the biological characteristics of the disease are more intrusive. In addition to LDH and platelet counts, the association between prognoses and such biological characteristics as chromosome pattern,19 cell markers,20 cell kinetics81 and agar colony growth22 have been clarified through several reports. Understanding of these factors, in conjunction with further technical refinements and multivariate analyses may lead to more accurate prognoses. A CKNOWLEDGMENT The authors are grateful to Dr. Haruhiko Sakurai for preparation of this manu script. REFERENCES 1. Wiernik, P. H. and Serpick, A. A.: Factors effecting remission and survival in adult acute nonlvmnhocvtic leukemia (ANLL). Medicine 49: , Gale, R. P : Advances in the treatment of acute myelogenous leukemia. N. Engl. J. Med. 300: , Kawashima, K., Suzuki, H., Yamada, K. et al.: Long-term survival in acute leukemia in Janan. Cancer 45: , Kimura, K.: Chemotherapy of acute leukemia with special reference to criteria for evaluation of therapeutic effect. In: Advances on Chemotherapy of Acute Leukemia. A Seminar on Chemotherapy of Acute Leukemia under the Japan-U.S. Cooperative Science Program, Sept., Bethesda, MD, 1965, Keating, M. J., Smith, T. L., Gehan, E. A. et al.: Factors related to iengtn or compiere rrmiscinn in adhilt acute leukemia. Cancer 45: , Keating, M. J., Smith, T. L., Gehan, F. A. et al.: A prognostic factor analysis for use in development of predictive models for response in adult acute leukemia. Cancer 50: , Armitage, J. 0., Burns, C. P., Dick, F. R. et al.: Results of aggressive cnemotnerapy for acute nonlymphoblastic leukemia in adults. Cancer Treat. Rep. 66: , Smith, T. L., Gehan, E. A., Keating, M. J, et al.: Prediction of remission in adult acute leukemia, development and testing of predictive models. Cancer 50: , Passe, S., Mike, V., Mertelsmann, R. et al.: Acute Nonlymphoblastic leukemia, prog nostic factors in adults with long-term follow-up. Cancer 50: , Suzuki, K., Kato, T., Yamada, K. et al.: Nippon Rinsho 39: , 1981 (in Japa nese Leimert, J. T., Burns, C. P., Wiltse, C. G. et al.: Prognostic influence of pretreatment characteristics in adult acute lymphoblastic leukemia. Blood 56: , Mertelsmann, R., Thaler, T., To, L. et al.: Morphological classification, response to therapy, and survival in 263 adult patients with acute nonlymphoblastic leukemia. Blood 56: , Keating, M. J.: Early identification of potentially cured patients with acute myelogenous leukemia-a recent challenge. In: Adult Leukemia 1 edited by Bloomfield, C. I. Martinus Nijhoff Pub., Hague/Boston/London, 1982, Bernard, J., Weil, M., Boiron, M. et al.: Acute promyelocytic leukemia: results of treat ment by Daunorubicin. Blood 41: , Keating, M. J., McCredie, K. B., Bodey, G. P. el al.: Improved prospects for long-term

7 Prognostic Factor in Adult Acute Leukemia 23 survival in adults with acute myelogenous leukemia. J. Amer. Med. Assoc. 248: , Gale, R. P., Zighelboim, J., Foon, K. A. et al.: Intensive chemoimmunotherapy in acute myelogenous leukemia (AML). Blood 52: 250, Bierman, H. R., Hill, B. R. and Emory, E.: Correlation of serum lactic dehydrogenase activity with clinical status of patients with cancer, lymphomas, and the leukemias. Cancer Res. 17: Brandt, L., Levan, G., Mitelman, F. et al.: Defective differentiation of megakaryocytes in acute mveloid leukemia. Acta. Med. Scand. 196: Sakurai, M. and Sandberg, A. A.: Prognosis of acute myeloblastic leukemia: chromosomal correlation. Blood 41: , Richie, E. R., Culbert, S. T., Sullivan, M. P. et at.: Complement receptor-positive, sheep erythrocyte receptor-negative lymphoblasts in childhood acute lymphocytic leukemia. Cancer Res. 38: , Hiddelmann, W., Buchner, T., Andreeff, M. et al.: Cell kinetics in acute leukemia-a critical reevaluation based on new data. Cancer 50: Moore, M. A., Spitzer, G., Williams, N. et al.: Agar culture studies in 127 cases of untreated acute leukemia: the prognostic value of reclassification of leukemia according to in vitro growth characteristics. Blood 44: