Japanese Journal of Clinical Oncology 31:382-387 (2001)
© 2001 Foundation for Promotion of Cancer Research
Retrospective Study of Hyponatremia in Gastric Cancer Patients Treated with a Combination Chemotherapy of 5-Fluorouracil and Cisplatin: a Possible Warning Sign of Severe Hematological Toxicities?
Division of Digestive Endoscopy and Gastrointestinal Oncology, National Cancer Center Hospital East, Kashiwa, Chiba, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Background: Some anti-neoplastic agents induce hyponatremia. The relationship between hyponatremia and other toxicities in gastric cancer patients treated with 5-fluorouracil and cisplatin (FP) was investigated retrospectively to clarify its clinical significance.
Methods: The subjects were 50 advanced gastric cancer patients treated with FP. Patients’ performance status, oral intake, nausea/vomiting, diarrhea, fever, urine volume, presence of ascites or pleural effusion, laboratory data and administration of diuretics, corticosteroid and contents and volume of hydration before and during the first 5 days after chemotherapy were reviewed.
Results: The serum sodium level decreased after initiation of chemotherapy in all patients and the lowest level (nadir) was most frequently observed on day 8 (range, days 2–14), which preceded hematological toxicities. In 10 patients (20%) the nadir of serum sodium was lower than 125 mEq/l. We classified these 10 patients as a low-sodium group and the others into a normal-sodium group. Six (60%) and seven (70%) of the 10 patients in the low-sodium group had complications with grade 3 or 4 leukopenia and thrombocytopenia, whereas only one (3%) and two (5%) were seen in the normal-sodium group (p < 0.0001). Stomatitis and diarrhea were also slightly more severe in the former than the latter group. With respect to sensitivity and probability, receiver operating characteristic curves showed the nadir (
or <125 mEq/l) of the serum sodium level was the best marker for both leukopenia and thrombocytopenia.
Conclusion: Hyponatremia after initiation of chemotherapy with FP may be a warning sign of subsequent severe hematological toxicity.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Chemotherapy has been demonstrated to provide a survival benefit for patients with unresectable or recurrent gastric cancer (1). Cisplatin (CDDP) is an active anti-neoplastic agent for various kinds of malignant diseases and some combination chemotherapies including cisplatin have been reported to be effective for gastric cancer (2–5). However, toxicities from these chemotherapies sometimes deteriorate the patients’ quality of life and treatment-related death is occasionally observed (6). The combination chemotherapy of 5-fluorouracil (5-FU) and CDDP (FP) is one of the most commonly used regimens worldwide (7–9). In a phase III trial by the Japan Clinical Oncology Group, continuous infusion of 5-FU (5-FUci) versus FP versus uracil and futrafur + mitomycin C, FP was more toxic than 5-FUci (10).
Patients with poor performance status (PS), organ dysfunctions and a history of prior chemotherapy and/or radiation therapy are well known to develop severe toxicities occasionally. However, it is difficult to predict this for patients in good condition, although some genetic changes have been reported to predispose toxicity (11–13). Hence it is very important to be able to predict or detect those patients who will develop severe toxicities in clinical situations.
It is well known that CDDP has gastrointestinal and nephro-, neuro- and hematological toxicities (2–10). In addition, CDDP modulates Na+-coupled uptake and membrane fluidity in renal proximal tubles (14) and it has been reported that some patients develop complications with hyponatremia during and after chemotherapy with regimens including CDDP (15). However, in clinical practice little attention is paid to hyponatremia.
To clarify the clinical significance of hyponatremia, we retrospectively investigated its incidence and grade and its relationship with other toxicities in gastric cancer patients treated with FP.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects
The subjects were 50 consecutive gastric cancer patients treated with FP, including six patients with adjuvant chemotherapy, at the National Cancer Center Hospital East between April 1994 and December 1995. According to the clinical practice of our institution, there were no definite eligibility criteria for the treatment with FP. However, in principle, FP was administered to patients with favorable medical conditions and not to those with massive ascites, poor performance status, active gastrointestinal bleeding, organ dysfunctions or other severe medical conditions. The treatment schedule using FP comprised protracted infusion of 5-FU (800 mg/m2/day) and 2-h infusion of CDDP (20 mg/m2/day) with adequate hydration on days 1–5, repeated every four weeks. Granisetron and in some patients a corticosteroid (decadron, 8 or 16 mg) were administered on day1 in order to reduce nausea and vomiting. Chemotherapy was conducted with patients’ informed consent.
Data Collection
Laboratory data were monitored at least once per week in all subjects. Administration of diuretics, decadron and the contents and volume of fluid administered during the initial 5 days after initiation of FP was monitored. Patients’ clinical status before and/or after chemotherapy, such as PS, oral intake, nausea/vomiting, diarrhea, fever, urine volume, presence of ascites or pleural effusion, was also reviewed. Hematological and non-hematological toxicities were evaluated according to the common toxicity criteria of the National Cancer Institute (website: http://ctep.info.nih.gov.). These clinical data during the first course were obtained from medical charts retrospectively.
Statistical Analysis
The relationship between patients’ status, symptoms, laboratory data before and/or after initiation of chemotherapy, administration of diuretics, decadron and grade 3 or 4 leukopenia and thrombocytopenia was investigated. The sensitivity and probability for these hematological toxicities were calculated and receiver operating characteristic (ROC) curves (16) analysis was performed to compare the utilities of these markers. The categories of these markers were as follows: gender, male/female; age /<60 years old; PS on ECOG scale, /<2; oral intake, /<50%; prior surgery, +/–; prior chemotherapy, +/–; ascites or pleural effusion, +/–; white blood cell count (WBC), /<4000 /mm3; serum level of albumin, /<3.5 g/dl; sodium, /<135 mEq/l (the lower limit of normal range of our institution is 138 mEq/l); bilirubin, /<1.5 mg/dl; aspartate aminotransferase (AST) and alanine aminotransferase (ALT), /<100 IU/dl; creatinine clearance, /<50 ml/min; vomiting and diarrhea grade 1, +/–; fever, /<37°C; administration of decadron, +/–; diuretics, +/–; and nadir of serum sodium level, /<125 mEq/l. Statistical analysis was performed with the chi-squared and Fisher’s exact tests and Student’s t-test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Serum Sodium Level After Initiation of FP
The serum sodium levels after initiation of chemotherapy decreased in all patients and the lowest level (nadir) of serum sodium was most frequently observed on day 8 (range, days 2–14) (Fig. 1A). Fig. 1B shows the distribution of the nadir of serum sodium, which ranged between 110 and 140 mEq/l. Although in most patients the nadir was
125 mEq/l, 10 patients (20%) developed <125 mEq/l hyponatremia. Fig. 1C shows the difference between serum sodium levels observed immediately before chemotherapy and those at the nadir. In 10 patients (20%) the serum sodium level fell by 15 mEq/l after chemotherapy. According to the nadir of serum sodium, we divided the subjects into a low-sodium group (<125 mEq/l, 10 patients) and a normal-sodium group (125 mEq/l, 40 patients).
|
Clinical Background of Low-sodium and Normal-sodium Groups
Table 1 shows patients’ characteristics before chemotherapy for the low-sodium and normal-sodium groups. Almost all patients showed PS 2 or better, with one patient in the normal-sodium group having a PS of 3. There were no remarkable differences in age, performance status, tumor extent, presence or absence of pleural effusion and ascites, history of prior chemotherapy or amounts of oral intake. Six (60%) of the 10 patients in the low-sodium group and 12 of the 40 (30%) patients in the normal-sodium group had undergone gastrectomy (p = 0.1381).
|
Table 2 shows laboratory data for the two groups before chemotherapy. There were no significant differences in liver and renal functions and serum albumin levels between the two groups. Five of the 10 (50%) patients in the low-sodium group had pretreatment serum sodium levels <135 mEq/l, with a minimum of 129 mEq/l, compared with five of the 40 (13%) in the normal-sodium group, with a minimum of 133 mEq/l (p = 0.0080). Six (60%) of the 10 patients in the low-sodium group had a WBC of <4000/mm3, with a minimum of 3400/mm3, compared with only two patients (5%) in the normal-sodium group, with a minimum of 2900/mm3 (p = 0.0110). In all patients platelet counts were >1 x 105/mm3 before the initiation of chemotherapy.
|
Factors Influencing the Serum Sodium Level Over 5 Days After Initiating FP
Table 3 shows the factors influencing the serum sodium level over 5 days after initiating chemotherapy. More patients in the low-sodium group (7/10, 70%) than in the normal-sodium group (13/40, 33%) showed complications with impairment of oral intake (p = 0.0673) and vomiting was observed more frequently in the former group (p = 0.1381). Although there were no differences in the use of diuretics between the two groups, decadron was used in more patients in the normal-sodium group. The total volume of hydration and the total amount of sodium administered were slightly higher in the low-sodium group than in the normal-sodium group (p = 0.0742) and the urine volume was slightly larger in the normal-sodium group than in the low-sodium group (p = 0.5046).
|
Hyponatremia and Toxicities
Leukopenia and thrombocytopenia were observed around 2 weeks after initiation of chemotherapy and the nadir of serum sodium had preceded these hematological toxicities in all patients. Table 4 shows hematological and non-hematological toxicities in the low- and normal-sodium groups. Six of the 10 patients (60%) in the low-sodium group developed grade 3 or 4 leukopenia, whereas only one patient (3%) did so in the normal-sodium group (p < 0.0001). More patients showed complications with grade 3 or 4 thrombocytopenia in the low-sodium group (seven patients, 70%) than in the normal-sodium group (two patients, 5%) (p < 0.0001). The incidences of grade 2 or higher stomatitis and diarrhea were slightly greater in the low-sodium group than the normal-sodium group. There were two treatment-related deaths from neutropenic sepsis, one in the first course and the other in the second course of chemotherapy; both patients were in the low-sodium group.
|
ROC Curves for Sensitivity and Probability
The ROC curves for markers and grade 3 or 4 leukopenia (Fig. 2A) and thrombocytopenia (Fig. 2B) were similar. Of the pretreatment factors, only WBC (
or <4000/mm3) and of the factors after initiation of chemotherapy, nadir of serum sodium ( or <125 mEq/l), oral intake ( or <50%) and administration of decadron were found to be predictors of grade 3 or 4 leukopenia. The nadir of serum sodium was the best marker in terms of sensitivity and probability. Thrombocytopenia was best predicted by the sodium nadir, with decadron and oral intake also being good markers.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Hyponatremia is a common adverse event in cancer patients and anti-neoplastic agents such as vincristine, vinblastine, cisplatin and cyclophosphamide are well known to induce hyponatremia (17). Williamson et al. reported that hyponatremia is one of the major toxicities after treatment with low-dose 5-FU and weekly CDDP in advanced gastric cancer patients (18). According to the NCI common toxicity criteria, hyponatremia levels <130 and 120 mEq/l are classified as grade 3 and 4, respectively. In this study, the cut-off level between the low- and normal-sodium groups was determined by ROC curves among 130, 125 and 120 mEq/l and differences in serum sodium level,
10 and 15 mEq/l between immediately before chemotherapy and nadir. A sodium nadir <125 mEq/l was the best predictor for hematological toxicities. For other pretreatment factors, cut-off levels were determined along with the eligibility criteria commonly adopted in many clinical trials. The serum sodium levels decreased after chemotherapy in all patients and 10 of 50 patients (20%) developed severe hyponatremia of <125 mEq/l, and the nadir of serum sodium was observed most frequently on day 8. In our institution laboratory data were usually monitored once per week. Closer monitoring might be able to detect hyponatoremia earlier and compensation for hyponatremia after its detection could reduce the incidence of severe hyponatremia. It seems that hyponatremia is common in gastric cancer patients treated with FP regardless of its degree, and sodium nadir preceded hematological toxicities.
The precise mechanism that induces hyponatremia was not determined. Because there were no remarkable differences in the incidences of vomiting, diarrhea and use of diuretics and sodium excretion into urine was within the normal range in one patient in the low-sodium group (data not shown), loss of sodium did not seem to be a major cause of hyponatremia. Of the pretreatment factors, more patients had received prior gastrectomy in the low-sodium group than the normal-sodium group. Impaired oral intake after chemotherapy might have influenced the serum sodium level after chemotherapy. However, in our institution the amount of sodium administered during the treatment of FP was usually around 150 mEq/day, which corresponds to 9 g of sodium chloride. Although diuretics were used in some patients, this is thought to be sufficient for all patients. Moreover, although the amounts of sodium administered in the low-sodium and the normal-sodium groups over the 5 days after initiation of chemotherapy were similar, the nadir of serum sodium was most frequently observed on day 8. Hence the absolute amount of sodium in the body was not considered a major cause of hyponatremia.
Cisplatin-containing regimens have been reported to induce hyponatremia in 4–10% of cases (17). The mechanism of CDDP action on the steady-state level of sodium is only partially known. It has been reported that the syndrome of inappropriate anti-diuretic hormone secretion (SIADH) is induced by CDDP (17). Since serum and urine osmolality, amount of sodium excreted in the urine and ADH level were not examined in this retrospective study, we could not clarify the precise mechanism of hyponatremia. Although the hydration volume was slightly larger in the low-sodium group than the normal-sodium group, the urine volume was slightly smaller in the former group. Some mechanism related to SIAHD might be a cause of hyponatremia.
Severe leukopenia and thrombocytopenia were observed more frequently in the low-sodium group (60 and 30%, respectively) than the normal-sodium group (3 and 5%, respectively). In a phase III study, the incidences of grade 3 or 4 leukopenia and thrombocytopenia were 25 and 19%, respectively, in 102 patients treated with FP (10). Hence the incidence of severe hematological toxicity in the low-sodium group is considered to be remarkably high. The ROC curves showed that, of the various pretreatment factors, only WBC could be a predictive marker of grade 3 or 4 leukopenia. However, the minimum WBC and platelet counts before chemotherapy in the low-sodium group were >3000/mm3 and >1 x 105/mm3, respectively. Hence it is difficult to predict which patients will suffer from severe hematological toxicities before initiation of chemotherapy. The nadir of serum sodium was the best marker of both leukopenia and thrombocytopenia. Moreover, treatment-related deaths were observed only in the low-sodium group. As far as we know, this is the first report of hyponatremia and hematological toxicities in chemotherapy involving CDDP. It is suggested that more attention should be paid to severe hematological toxicity when hyponatremia is observed.
The precise mechanism by which hematological toxicity developed in the low-sodium group is not known. Among the pretreatment factors, there were more patients with low WBC, prior surgery and impairment of oral intake before and after chemotherapy in the low-sodium group than the normal-sodium group and the ROC curves showed that impairment of oral intake after initiation of FP was also a marker of severe hematological toxicities. A patient’s general status, such as oral intake, might be related to hematological toxicities. Because decadron was used in more patients in the normal-sodium group than the low-sodium group, it might be related to hyponatremia and/or severe hematological toxicities to some degree. However, it is not known whether decadron could prevent hyponatoremia and/or severe hematological toxicities. Some authors have reported a relationship between degradation of 5-FU and its toxicities (11,12). Because the incidences of 5-FU-related non-hematological toxicities such as grade 2 or greater stomatitis and diarrhea were slightly higher in the low-sodium group, degradation of 5-FU might have been altered in this group.
In conclusion, although there may be some unknown mechanisms causing severe hematological toxicities, hyponatremia, especially if <125 mEq/l, can be a warning sign of severe hematological toxicity after initiation chemotherapy with FP in gastric cancer patients.
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Narikazu Boku, Division of Digestive Endoscopy and Gastrointestinal Oncology, National Cancer Center Hospital East, 6–5–1, Kashiwanoha, Kashiwa, Chiba 277-8577, Japan. E-mail: nboku@east.ncc.go.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 Murad AM, Santiago FF, Petroianu A, Rocha PR, Rodrigues MA, Rausch M. Modified therapy with 5-fluorouracil, doxorubicin and methotrexate in advanced gastric cancer. Cancer 1993;72:37–41.[Web of Science][Medline]
2 Moertel CG, Rubin J, O’Connell MJ, Schutt AJ, Wieand HS. A phase II study of combined 5-fluorouracil, doxorubicin and cisplatin in the treatment of advanced upper gastrointestinal adenocarcinomas. J Clin Oncol 1986;4:1053–7.
3 Preusser P, Wilke H, Achterrath W, Fink U, Lenaz L, Heinicke A, et al. Phase II study with the combination etoposide, doxorubicin and cisplatin in advanced measurable gastric cancer. J Clin Oncol 1989;7:1310–7.[Abstract]
4 Findlay M, Cunningham D, Norman A, Mansi J, Nicolson M, Hickish T, et al. A phase II study in advanced gastric cancer using epirubicin and cisplatin in combination with continuous 5-fluorouracil (ECF). Ann Oncol 1994;5:609–16.
5 Boku N, Ohtsu A, Shimada Y, Shirao K, Seki S, Saito H, et al. A phase II study of a combination of irinotecan and cisplatin against metastatic gastric cancer. J Clin Oncol 1999;17:319–23.
6 Kelsen D, Atiq OT, Saltz L, Niedzwiecki D, Ginn D, Chapman D, et al. FAMTX versus etoposide, doxorubicin and cisplatin: a randomized trial in gastric cancer. J Clin Oncol 1992;10:541–8.
7 Ohtsu A, Shimada Y, Yoshida S, Saito H, Seki S, Morise K, et al. Phase II study of protracted infusional 5-fluorouracil combined with cisplatinum for advanced gastric cancer: report from the Japan Clinical Oncology Group (JCOG). Eur J Cancer 1994;30A:2091–3.
8 Rougier P, Ducreux M, Mahjoubi M, Pignon JP, Bellefqih S, Oliveira J, et al. Efficacy of combined 5-fluorouracil and cisplatinum in advanced gastric carcinomas: a phase II trial with prognostic factor analysis. Eur J Cancer 1994;30A:1263–9.
9 Kim NK, Park YS, Heo DS, Suh C, Kim SY, Park KC, et al. A phase III randomized study of 5-fluorouracil and cisplatin versus 5-fluorouracil, doxorubicin and mitomycin C versus 5-fluorouracil alone in the treatment of advanced gastric cancer. Cancer 1993;71:3813–8.[Web of Science][Medline]
10 Shimada Y, Shirao K, Ohtsu A, Hyodo I, Saito H, Yamamichi N, et al. Phase III study of UFT + MMC versus 5-FU + CDDP versus 5-FU alone in patients with advanced gastric cancer. JCOG study 9205 (abstract). Proc Am Sci Clin Oncol 1999;1043
11 Gonzalez FJ, Fernandez-Salguero P. Diagnostic analysis, clinical importance and molecular basis of dihydropyrimidine dehydrogenase deficiency (review). Trends Pharmacol Sci 1995;16:325–7.[Medline]
12 Diasio RB, Beavers TL, Carpenter JT. Familial deficiency of dihydropyrimidine dehydrogenase. Biochemical basis for familial pyrimidinemia and severe 5-fluorouracil-induced toxicity. J Clin Invest 1988;81:47–51.
13 Krynetski EY, Evans WE. Pharmacogenetics of cancer therapy: getting personal (review). Am J Hum Genet 1998;63:11–6.[Web of Science][Medline]
14 Courjault-Gautier F, Le Grimellec C, Giocondi MC, Toutain HJ. Modulation of sodium-coupled uptake and membrane fluidity by cisplatin in renal proximal tubular cells in primary culture and brush-border membrane vesicles. Kidney Int 1995;47:1048–56.[Web of Science][Medline]
15 el Weshi A, Thieblemont C, Cottin V, Barbet N, Catimel G. Cisplatin-induced hyponatremia and renal sodium wasting. Acta Oncol 1995;34:264–5.[Web of Science][Medline]
16 Metz CE. Basis principles of ROC analysis. Semin Nucl Med 1978;8:283–98.[Web of Science][Medline]
17 Berghmans T. Hyponatremia related to medical anticancer treatment (review). Supportive Care Cancer 1996;4:341–50.
18 Williamson SK, Tangen CM, Maddox AM, Spiridonidis CH, Macdonald JS. Phase II evaluation of low-dose continuous 5-fluorouracil and weekly cisplatin in advanced adenocarcinoma of the stomach. Am J Clin Oncol 1995;18:484–7.[Web of Science][Medline]
Received February 19, 2001; accepted April 25, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||