Cushing's Syndrome Associated With Ectopic Corticotropin Production and Small-Cell Lung Cancer By Frances A. Shepherd, Jennifer Laskey, William K. Evans, Paul E. Goss, Else Johansen, and Firouz Khamsi Purpose: This study was undertaken to review the clinical and laboratory features and response to treatment of patients with Cushing's syndrome associated with ectopic corticotropin (adrenocorticotropic hormone; ACTH) production and small-cell lung cancer (SCLC). Patients and Methods: We undertook a retrospective chart review of 545 patients with SCLC seen at Toronto General Hospital between 1980 and 1990 and identified 23 patients (4.5%) with Cushing's syndrome and ectopic ACTH production. Results: There were 17 male and six female patients, with a median age of 60 years. The syndrome was diagnosed at the time of initial presentation of SCLC in 13 patients and at relapse in 10 patients. Seven patients had limited disease and 16 had extensive disease at their initial diagnosis of SCLC, but 20 of 23 had extensive disease at the time of diagnosis of Cushing's syndrome. Ten patients had bone marrow involvement. The most frequent physical findings included edema (83%) and
S
MALL-CELL lung cancer (SCLC) accounts for 20% to 25% of all bronchogenic neoplasms and is distinguishable clinically and biologically from other primary lung cancers by its rapid doubling time, its propensity for early hematogenous dissemination, and its excellent responsiveness to combination chemotherapy. SCLC is also unusual because of its frequent association with paraneoplastic syndromes, which occur in approximately 20% of patients at presentation.' The most frequent and best understood syndromes are those that result from ectopic hormone secretion by the tumor. Although a large variety of peptide hormones have been shown to be secreted by SCLC cells, ectopic production of antidiuretic hormone (ADH) and corticotropin (adrenocorticotropic hormone; ACTH) are seen most frequently.' The production of such hormones may or may not be associated with overt clinical syndromes, depending on the functional activity and level of the ectopic hormone produced. Clinical Cushing's syndrome secondary to ectopic ACTH production is uncommon, occurring in approximately 5% of all SCLC patients.' However, biochemical evidence of hypercortisolism can be detected in up to 50% of patients,2'3 and it is possible to identify immunoreactive ACTH in almost all SCLC tissue extracts.4 The relationship between ectopic ACTH production and tumor burden has been difficult to study because of the small number of patients with clinical evidence of
proximal myopathy (61%). All patients had elevated plasma and urinary free cortisol levels; 22 had a hypokalemic alkalosis, and 13 had hyperglycemia. Only one patient had a normal ACTH level. The response rate (complete plus partial) to chemotherapy for patients who had the syndrome diagnosed at initial presentation of SCLC was only 46%, and their median survival was only 3.57 months. Only two patients achieved complete normalization of all hormone parameters, and neither experienced hormone relapse at the time of SCLC relapse. Complications of therapy included gastrointestinal (GI) ulceration (six patients), GI bleeding (four), perforation of a duodenal ulcer (one), pneumonia (10), septic shock (three), and fungal infections (five). Conclusion: Ectopic ACTH production is associated with a low response to chemotherapy, short survival, and a high rate of complication to therapy. J Clin Oncol 10:21-27. © 1992 by American Society of Clinical Oncology. the syndrome. One point of apparent consensus is that patterns of ACTH secretion vary considerably. Importantly, plasma ACTH levels do not necessarily correlate directly with tumor burden, and relapse of a hormonesecreting tumor after response is not always accompa-
nied by resumption of hormone secretion. As a result, 56 ACTH can not be used as a tumor marker. , The impact of ectopic ACTH secretion associated
with clinical Cushing's syndrome on disease course and overall survival is unclear. Earlier studies suggested that
ectopic ACTH production might be associated with 37 8 more extensive disease and a worse prognosis, " ' but the
total number of patients reported in these studies is
small. Furthermore, the impact of aggressive combination chemotherapy and new and more effective means of
treating the metabolic abnormalities caused by hypercortisolism have not been assessed. In an attempt to resolve some of these issues, we
From the Division of Hematology and Oncology, Department of Medicine, Toronto Hospital, Toronto General Division, Toronto; and the University of Toronto, Toronto, Canada. Submitted May 22, 1991; acceptedJuly 26, 1991. Address reprintrequests to FrancesA. Shepherd, MD, M/L W2-035, Toronto General Hospital, 200 Elizabeth St, Toronto, Ontario M5G 2C4, Canada. © 1992 by American Society of ClinicalOncology. 0732-183X/9211001-0022$3.00/0
Journalof Clinical Oncology, Vol 10, No 1 (January), 1992: pp 21-27
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
21
22
SHEPHERD ET AL
reviewed the records of all patients with SCLC seen at our institution over a 10-year period to identify patients with ectopic ACTH production associated with clinical Cushing's syndrome. We report here the results of this review. PATIENTS AND METHODS We undertook a 10-year (1980 to 1990) retrospective chart review of 545 patients with histologically or cytologically confirmed SCLC treated at the Toronto General Hospital. From this review, 23 patients with ectopic ACTH production were identified. Patients were considered to have an ectopic ACTH syndrome when there were clinical signs and symptoms of excess corticosteroid production together with at least two of the following conditions: hypokalemia defined as a serum potassium level less than 3.0 mmol/L, elevated plasma ACTH (> 22 pmol/L), elevated plasma cortisol (> 660 nmol/L) with loss of diurnal variation and/or lack of suppressibility by dexamethasone, or an elevated 24-hour urinary free cortisol level (> 400 nmol/d). Pretreatment clinical staging was performed for all patients and included a complete history and physical examination, determination of Eastern Cooperative Oncology Group (ECOG) performance status, a complete blood cell count, and a biochemical profile. Radiologic staging procedures included a chest x-ray, a radionuclide bone scan, an ultrasound or computed tomographic (CT) scan of the abdomen, either a radionuclide or CT scan of the brain, and a bone marrow aspirate and trephine biopsy. Based on the results of the above investigations, disease was classified as either limited (disease confined to the thorax plus the ipsilateral supraclavicular fossa) or extensive. All patients were treated with combination chemotherapy using either cyclophosphamide, doxorubicin, and vincristine (CAV) or etoposide (VP-16)-cisplatin (VP-CP). CAV consisted of cyclophosphamide 1,000 mg/m2, doxorubicin 50 mg/m2, and vincristine 2 mg all given intravenously at 3-week intervals, usually for six cycles. VP-CP consisted of VP-16 100 mg/m 2 and cisplatin 25 mg/m', both given intravenously daily for 3 days at 3- to 4-week intervals for six cycles. Standard SCLC tumor response criteria were used. The response to ectopic hormone therapy was evaluated by measurement of posttreatment plasma ACTH, serum and urinary free cortisol, potassium, and blood sugar levels. A complete hormone response was defined as normalization of all abnormal biochemical parameters related to the ectopic hormone secretion. Partial response was defined as a 50% reduction in the level of the abnormal biochemical parameter that was above the upper limit of normal. Survival time was measured from the date of diagnosis of SCLC to the date of death. Actuarial survival curves were prepared by means of the Kaplan-Meier method,' and comparisons of survival were performed using a Wilcoxon log-rank sum test.'o
RESULTS We reviewed the records of 545 patients with a diagnosis of SCLC seen at the Toronto General Hospital between 1980 and 1990, and identified 23 patients
(4.5%) who had clinically recognizable hypercortisolism. The characteristics of these 23 patients are listed in Table 1. There were 17 male and six female patients, with a median age of 60 years (range, 43 to 77 years). At the time of initial diagnosis of lung cancer, seven
Table 1. Patient Characteristics
No. of patients Median age, years (range) Male:female ratio Performance status (ECOG) 0, 1 2 3,4
Limited Disease
Extensive Disease
Total
7 59 (47-69) 4:2
16 60 (43-77) 13:4
23 60 (43-77) 17:6
6 0 1
12 1 3
18 1 4
5 2
14 2
19 4
3 4*
10 6
13 10
Paraneoplastic syndrome
Ectopic ACTH only Ectopic ACTH and SIADH Ectopic hormone first noted at
Presentation Relapse Abbreviation: SIADH, syndrome
of inappropriate antidiuretic hor-
mone. *Three patients who, at their initial diagnosis of SCLC, had limited disease had extensive disease at relapse when the ectopic ACTH first appeared.
patients (30%) had limited disease and 16 (70%) had extensive disease. Of the seven patients with limited disease, three had ectopic ACTH production at presentation, and four developed the syndrome at relapse when there was also evidence of hematogenous metastases. Of the 16 patients with extensive disease at presentation, 10 had ectopic ACTH production at presentation, and in six, it developed only at relapse. Two of the patients whose ectopic hormone syndrome was diagnosed at relapse had demonstrated transient hypokalemia earlier in the course of their disease, but the syndrome was only documented at the time of relapse. In total, 13 of the 23 patients had evidence of ectopic ACTH production at the time of their initial diagnosis of SCLC, 10 developed the syndrome at relapse, and 20 had extensive disease when the syndrome was first identified. In the 20 patients who had evidence of hematogenous metastases at the time of onset of ectopic ACTH production, liver metastases were present in 15, bone metastases in 13, brain metastases in seven, and bone marrow involvement was documented in 10 patients (Table 2). In addition, of the three patients who had limited disease and ectopic ACTH at presentation, two had thrombocytopenia and a leukoerythroblastic blood film suggestive of bone marrow involvement, but both had normal bone marrow biopsies. One extensivedisease patient had thrombocytopenia at relapse, but a bone marrow biopsy was not performed to confirm metastatic involvement. In total, 13 patients (56%) probably had bone marrow metastases. All patients had either normal or slightly elevated baseline WBC and absolute granulocyte counts at presentation. The me-
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
23
ECTOPIC ACTH AND SCLC Table 2. Distribution of Metastases
Table 4. Laboratory Findings
Site of Metastases
Frequency (%)
Laboratory Parameter
No. Tested
Liver Bone Bone marrow Confirmed by biopsy Thrombocytopenia and leukoerythroblastic blood films CNS Adrenal gland Other
15 (75) 13(65)
Elevated plasma cortisol level Loss of cortisol, diurnal variation Elevated 24-hour urinary free cortisol Elevated plasma ACTH level Hypokalemia Hyperglycemia Hyponotremia
22 20 14 16 23 22 23
10 (50) 3* 7 (35) 4 (20) 2 (10)
*Two patients had limited disease and normal bone marrow biopsies, and one patient with extensive disease did not have a bone marrow biopsy performed.
dian pretreatment platelet count was 157 x 109/L (range, 29 x 109/L to 430 x 109/L), and eight patients
No. Abnormal (%) 22 20 14 15 22 13 4
(100) (100) (100) (94) (96) (59) (17)
sponse, one achieved partial response, and one patient died early. Only one of the 10 patients with extensive disease achieved complete response, three a partial response, and six had progressive disease. Only eight of the 10 patients received chemotherapy at the time of relapse, and three (30%) achieved partial response.
had counts less than 150 x 109/L.
The clinical features of Cushing's syndrome seen in the 23 patients are listed in Table 3. The most common features included peripheral edema (83%), proximal myopathy (61%), and moon facies (52%). Many of the other physical findings associated with the classical syndrome were also seen, but to a lesser extent. The relevant laboratory findings are listed in Table 4. All patients tested had elevated plasma cortisol levels with loss of diurnal variation, as well as elevated 24-hour urinary free cortisol levels. The plasma ACTH was elevated in 94% of patients, and 96% had hypokalemia associated with a metabolic alkalosis. Hyperglycemia was present in 59% of patients. Four patients had concurrent ectopic ACTH and inappropriate ADH production. Tumor Response The impact of combination chemotherapy on both the tumor burden and ectopic ACTH production is shown in Tables 5 and 6. The overall response rate for patients who had ectopic ACTH syndrome at the time of initial presentation of their SCLC was 46%. One of the three patients with limited disease achieved complete reTable 3. Clinical Features of Cushing's Syndrome Finding Peripheral edema Proximal myopathy Moon facies Buffalo hump Truncal obesity Hyperpigmentation Psychosis Hypertension Ecchymoses Hirsutism Striae
Frequency (%) 19 14 12 8 8 5 5 5 4 1 1
(83) (61) (52) (35) (35) (22) (22) (22) (17) (4) (4)
Hormone Response Of the 13 patients who had ectopic hormone production at the initial presentation of SCLC, only two had a complete response of their ectopic hormone production with normalization of all parameters measured. In neither case did the ectopic hormone syndrome recur at relapse. Seven patients had a partial response of their ectopic ACTH production to antitumor therapy with a reduction of ACTH and serum or urinary cortisol levels, but not to normal. Two patients experienced progressive disease with a progressive increase in their serum cortisol levels in spite of chemotherapy. Two patients were not assessable: one died prematurely due to pneumonia, and in another, postchemotherapy ACTH and cortisol levels were not measured. Of the ten patients who had their ectopic hormone production diagnosed at relapse, there were no complete responses. Five patients achieved a partial hormone response: two after treatment with chemotherapy only, two who were Table 5. Response to Combination Chemotherapy Response by Time of Presentation of Ectopic
Limited
Extensive
ACTH Production
Disease (%)
Disease (%)
Total (%)
3 1 (33) 1 (33) 0 1 (33)
10 1 (10) 3 (30) 6 (60) 0
13 2 (15) 4(31) 6 (46) 1 (8)
At initial presentation of SCLC No. of patients CR PR PD Early death At relapse of SCLC No. of patients No treatment CR PR PD
10 2 (20) 0 3 (30) 5 (50)
Abbreviations: CR, complete response; PR, partial progressive disease.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
response; PD,
24
SHEPHERD ET AL
Table 6. Summary of Treatment and Response Time of Diagnosis
Stage of SCLC
Response of Tumor
of Ectopic ACTH Production
(at presentation/ at relapse)
to Chemotherapy (primary/relapse)
At initial presentation of SCLC
L
Serum
Urinary Free
PR
PR
NT
NT
N
CR
PR
NT
4
NT
NA PR
NT ? PR
NT NA
NT NT
NT NT
PR
NE
NT
NT
NT
CR
CR
N
NT
NT
PD
PD
NC
T
NT
PD
PR
1
1
NT
PD
PR
PR
CR
N
N
NT
PD
PD
NT
1
NT
PD
PR
N
4
NT
E
PD
PR
4
4
N
E
PR/PR
PR
1
1
NC
L/E
PD/PR
Stable
NC
NT
NC
E
PR/PD
PD
N
NC
T
E
PR/PD
PD
4
T 4
E
CR/PD
PR
NT
NC
4
L/E
CR/PR
PR
NT
4
NT
E
PR/NA
PR
NT
NT
N
L/E
CR/PD
NE
NT
NT
NT
E
PD/NA
PR
NC
N
,
L/E
NA/PD
NE
NT
NT
NT
L E
E
E
E
At relapse of SCLC
Cortisol Cortisol
Response of
Syndrome to Therapy ACTH
$
Comments
K+ N: K+ supplementation continued N: treatment no longer required NT N: treatment no longer required N: never had hypokalemia N: treatment no longer required N: K+ supplementation continued N: K+ supplementation and amiloride continued N: K+ supplementation and spironolactone continued N: K+ supplementation stopped N: K+ supplementation continued N: K+ supplementation and spironolactone continued N: K+ supplementation continued N: K+ supplementation and spironolactone continued N: spironolactone continued N: K+ supplementation and spironolactone continued Fluctuating: K+ supplementation and spironolactone continued N: K+ supplementation and spironolactone N: treatment no longer required N: K+ supplementation continued N: spironolactone continued Fluctuating: K+ supplementation continued N: K+ supplementation continued
Early death
No return of ectopic ACTH at relapse
Treated with ketoconazole Treated with ketoconazole No return of ectopic ACTH at relapse, K+ stopped
Treated with ketoconazole
Treated with ketoconazole No response to ketoconazole or aminoglutethimide Treated with ketoconazole
Treated with ketoconazole
Treated with ketoconazole
Abbreviations: L, limited; E, extensive; NA, not applicable; NT, not tested; NC, no change; N, normal; K+, potassium; T , increased;
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1,,decreased.
25
ECTOPIC ACTH AND SCLC
treated with both chemotherapy and ketoconazole, and one who was treated with ketoconazole alone. One patient had stable hormone levels, two had progressive disease, and two died with progressive malignancy before hormone levels could be reevaluated. Ketoconazole was used in the treatment of eight patients. Only in one patient was ketoconazole given without concurrent chemotherapy, and a partial hormone response was achieved. Two of the remaining seven patients demonstrated no response, and one died before repeat hormone testing was undertaken. Four patients had partial hormone response even though their tumors did not respond clinically to chemotherapy. These patients also had some improvement in their ACTH levels, which suggests that both chemotherapy and ketoconazole may have contributed to the reduction in serum and urinary cortisol levels.
a,
o
-. m
0
Survival
20
10
The overall survival for the 23 patients is shown in Fig 1. The median survival of the whole group was 6.23 months. Within this group, patients with ectopic ACTH at initial presentation of SCLC had a significantly shorter survival (median, 3.57 months) compared with those with ectopic hormone diagnosed at relapse who had a median survival of 10.7 months from the date of diagnosis of their SCLC (P = .0020). This was true both for limited- and extensive-disease patients. The median survival of the limited-disease patients overall was only 27 weeks (range, 1 to 75 weeks), and of extensive-disease patients, 28 weeks (range, 3 to 87 weeks). Limited- and extensive-disease patients who had ectopic hormone secretion diagnosed at their initial presentation of SCLC had median survivals of only 9 and 18 weeks, respectively. Toxicity Significant hematologic toxicity was encountered after chemotherapy. The median nadir granulocyte count was only 0.24 x 109/L (range, 0 to 1.8 x 109/L), and six
patients had nadir granulocyte counts of 0. The median nadir platelet count was 33 x 109/L (range, 10 x 109/L to 306 x 109/L), and five patients had counts less than 20 x 109/L. A high rate of complication was seen during treatment in this series (Table 7). Six patients had gastrointestinal (GI) ulceration associated with upper GI bleeding in four patients and perforation of a duodenal ulcer in one. Serious infection occurred in 10 patients. Ten patients had pneumonia, and three experienced episodes of septic shock. Fungal infections were seen in five patients and included disseminated candidi-
TIME IN MONTHS Fig 1. A comparison of the survival from the date of diagnosis of SCLC for patients with ectopic ACTH diagnosed at presentation (----; n = 13; median survival, 3.57 months) or at relapse (; n = 10; median survival, 10.7 months) (P = .0020). (----) All patients; n = 23; median survival, 6.23 months.
asis in two, superficial candidiasis in two, and pulmonary aspergillosis in one. One patient each suffered massive hemoptysis, hemothorax, and gangrene of the foot. DISCUSSION
Although Cushing's syndrome associated with the ectopic production of ACTH has been reported in association with a variety of malignant tumors, it is seen most frequently in patients with SCLC.' Even in SCLC, the clinical syndrome is rare and is seen in 5% of patients or less. In our review, which was limited to Table 7. Complications During Treatment Complication
No. of Patients
Gastrointestinal ulceration Bleeding Perforation Infection Pneumonia Septic shock Disseminated candidiasis Superficial candidiasis Aspergillosis Massive hemoptysis Bullous impetigo Gangrene of the foot Hemothorax
6 4 1 10 10 3 2 2 1 1 1 1 1
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
26
SHEPHERD ET AL
patients with clinical manifestations of hypercortisolism, 23 of 545 patients (4.5%) were found to have a diagnosis of Cushing's syndrome, an incidence similar to that reported previously. Biochemical evidence of ACTH production is seen much more frequently. When radioimmunoassays are used, ACTH can be detected in virtually 100% of SCLC tissue extracts,4 and elevated plasma levels have ',een reported in one quarter to one third of patients.2 5 Bondy and Gilby3 assessed the endocrine function of 106 untreated patients with SCLC and found that 77% of their patients had abnormalities in at least one test that suggested derangement of adrenocortical function. The median age and male:female ratio of the patients in our study did not differ from that of SCLC overall, and most patients had a good performance status at the time of diagnosis. The most common clinical findings included peripheral edema, proximal myopathy, and moon facies, which were all seen in more than half the patients. Several authors have observed that classical "Cushingoid" features may be absent in patients with SCLC and ectopic ACTH, and that, as in our series, rapidly progressive weight loss, myopathy, edema, and hypertension are often the most prominent features of the syndrome." This may be due to the fact that the patients do not live long enough to develop the classical signs. Hypokalemia, often in association with a mild metabolic alkalosis, was seen in all but one of the patients in our series. In fact, it was usually the finding of a low serum potassium level that suggested the presence of ectopic ACTH secretion and led to subsequent confirmatory tests. Howlett et a11 2 have reported that severe hypokalemia occurs more frequently in ectopic ACTH syndromes than in other causes of Cushing's syndrome. All patients tested had elevated plasma cortisol levels with loss of diurnal variation, and elevated 24-hour urinary free cortisol levels. Only one patient who had markedly elevated serum and urinary cortisols had a normal ACTH level. Normal plasma ACTH levels in the presence of abnormal corticosteroid concentrations have been reported previously by Bondy and Gilby3 and may possibly be explained by the production of a functionally active but incomplete ACTH molecule that is not detected by standard radioimmunoassays. The incidence in our series of either new onset diabetes, or worsening of a previously diagnosed diabetic state was also similar to that seen by Bondy and Gilby, who reported abnormal glucose tolerance tests in 56% of the patients studied in their series. All patients in our series who presented initially with both SCLC and ectopic ACTH syndrome were treated
with combination chemotherapy. Only two patients (one limited and one extensive) achieved complete responses, and four patients (one limited and three extensive) achieved partial responses for an overall response rate of only 46%. The relationship between ectopic ACTH production and tumor behavior is not clearly defined in the literature because of the small number of patients in the reported series to date. Although early series suggested that SCLC associated with Cushing's syndrome might be associated with a worse prognosis, many patients in these series were not treated with the aggressive chemotherapy combinations that are current standard therapy. Our low response rate despite chemotherapy treatment with either CAV or VP-CP suggests that the presence of ectopic ACTH secretion does confer an adverse prognosis and that SCLC associated with ectopic ACTH production may be inherently more resistant to chemotherapy than other SCLC tumors. Cell culture and in vitro sensitivity testing could provide interesting data in this regard. Neither of the two patients who had complete normalization of their hormone levels had a return of their ectopic hormone syndrome at the time of relapse of their SCLC. This phenomenon has been reported previously for both ectopic ACTH and ADH secretion,5 as well as for other 1 paraneoplastic syndromes associated with SCLC." The median survival for the 13 patients with ectopic hormone secretion at presentation was less than 4 months and was significantly lower than that seen in patients who developed their syndrome only at the time of relapse. Abeloff et al8 reported a similar experience. The median survival from the time of diagnosis of SCLC in their series was only 7 months, and no patient survived 1 year. Similarly, in a small series reported by Lokich, 7 three of four patients survived 3 months or less. A closer examination of the cause of death in our series suggests that the reason for such poor survival may be multifactorial. Our review suggests that these patients have a much higher rate of complications during therapy than that seen in patients with SCLC not associated with an ectopic hormone syndrome. Forty-three percent of our patients experienced severe infections, and in all cases, the infection contributed significantly to the cause of death. Furthermore, three patients had invasive fungal infections with pulmonary involvement in all cases. Six patients had GI ulceration, four of these patients had active upper GI bleeding, and one suffered a perforation of a duodenal ulcer. An increased susceptibility to infection, particularly fungal infection, and GI ulceration are all recognized complications of hypercortisolism, whether on a pathologic or iatrogenic basis. However, it interesting that such a high incidence of these
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
27
ECTOPIC ACTH AND SCLC
complications was seen in our patient population, who all had hypercortisolism of fairly brief duration. Abeloff et al' reported pulmonary infection and sepsis in two of five patients in their series, and Lokich' reported pulmonary infiltrates possibly due to Pneumocystis carinii or other fungal organisms in two of four patients. The infectious and hemorrhagic complications after therapy may have resulted, in part, from severe posttreatment neutropenia and thrombocytopenia. More than half of our patients had bone marrow involvement by tumor. Ten were confirmed by biopsy, and three others had significant pretreatment thrombocytopenia and leukoerythroblastic blood films. Only one other review has suggested an association between bone marrow metastases and paraneoplastic syndromes. De la Monte et al"4 identified bone marrow involvement in 76% of 28 patients with paraneoplastic syndromes, which included ectopic ACTH secretion as well as other syndromes associated with SCLC. In a previous review, we identified bone marrow involvement in only 67 of 403 patients with SCLC (16.6%) treated in three consecutive University of Toronto lung cancer study protocols."5 Therefore, we believe that the level of bone marrow involvement seen in this series is unusually high, and we question whether there is perhaps an association between bone marrow involvement and ectopic ACTH secretion. In addition to chemotherapy, eight patients received specific therapy to block adrenocorticosteroid produc-
tion. Eight patients were treated with ketoconazole, and one patient received aminoglutethimide after not responding to ketoconazole. Ketoconazole is an imidazole derivative which blocks corticosteroid production in vitro and in vivo by inhibition of 14-demethylation of cholesterol.'" It has been used successfully in the treatment of ectopic ACTH secretion in association with SCLC'7 as well as other tumors.' Four patients achieved partial remission of their ectopic hormone syndrome despite progression of malignant disease while on chemotherapy. Two patients had progression of their malignant disease associated with worsening of their hormone levels. One patient who did not receive chemotherapy at relapse had a partial response in hormone levels during treatment with ketoconazole alone, and one patient died early before hormone testing could be repeated. We believe that ketoconazole contributed to control of hormone levels in five of the eight patients treated with this medication. In summary, clinical Cushing's syndrome secondary to ectopic ACTH secretion occurs in approximately 5% of patients with SCLC. It is associated with a low response to combination chemotherapy and a high rate of complication during treatment. Although corticosteroid levels may be controlled by the use of corticosteroid blocking agents such as ketoconazole, survival remains poor for this group of patients.
REFERENCES 1. Bunn PA, Ridgway EC: Parancoplastic syndromes, in DeVita V, Hellman S, Rosenberg SA (eds): Cancer Principles and Practice. Philadelphia, PA, Lippincott, 1989. pp 1896-1940 2. Hansen M. Hansen H, Hirsch F, et al: Hormonal polypeptides and amine metabolites in small cell carcinoma of the lung with special reference to stage and subtypes. Cancer 45:1432-1437, 1980 3. Bondy PK, Gilby ED: Endocrine function in small cell undifferentiated carcinoma of the lung. Cancer 50:2147-2153, 1982 4. Gewirtz G, Yalow R: Ectopic ACTH production in carcinoma of the lung. J Clin Invest 53:1022-1032, 1974 5. Gropp C, Havemann K, Scheuer A: Ectopic hormones in lung cancer patients at diagnosis and during therapy. Cancer 46:347354, 1980 6. Hansen M, Hammer M, Hummer L: ACTH, ADH and calcetonin as markers of response and relapse in small-cell carcinoma of the lung. Cancer 46:2062-2067, 1980 7. Lokich JJ: The frequency and clinical biology of the ectopic hormone syndromes of small cell carcinoma. Cancer 50:2111-2114, 1982 8. Abeloff M, Trump D. Baylin S: Ectopic adrenocorticotrophic (ACTH) syndrome and small cell carcinoma of the lung-Assessment of clinical implications in patients on combination chemotherapy. Cancer 48:1082-1087, 1981 9. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
10. Gehan EA: A generalized Wilcoxon test for comparing arbitrarily. Singly censored samples. Biometrika 52:203-223. 1982 11. Howlett TA, Rees LH, Besser GM: Cushing's syndrome. Clin Endocrinol Metabol 14:911-945, 1985 12. Howlett TA, Drury PL, Doniach PI, et al: Diagnosis and management of ACTH-dependent Cushing's syndrome: Comparison of the features in ectopic and pituitary ACTH production. Clin Endocrinol 24:699-713, 1986 13. Clamon G, Evans W, Shepherd F, et al: Myasthenic syndrome and small cell cancer of lung. Variable response to antineoplastic therapy. Arch Intern Med 144:99-100, 1984 14. De la Monte S, Hutchins G, Moore G: Paraneoplastic syndromes and constitutional symptoms in prediction of metastatic behavior of small cell carcinoma of the lung. Am J Med 77:851-856, 1984 15. Campling B, Ouirt I, DeBoer G, et al: Is bone marrow examination in small cell lung cancer really necessary? Ann Intern Med 105:508-512, 1986 16. Pont A, Williams P, Ltnose D, et al: Ketoconazole blocks adrenal steroid synthesis. Ann Intern Med 97:370-372, 1982 17. Shepherd F, Hoffert B, Evans W, et al: Ketoconazole: Use in the treatment of ectopic adrenocorticotropic hormone production and Cushing's syndrome in small-cell lung cance . Arch Intern Med 145:863-864, 1985 18. Soniro N: The use of ketoconazole as an inhibitor of steroid production. N Engl J Med 317:812-818. 1987
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
S
MALL-CELL lung cancer (SCLC) accounts for 20% to 25% of all bronchogenic neoplasms and is distinguishable clinically and biologically from other primary lung cancers by its rapid doubling time, its propensity for early hematogenous dissemination, and its excellent responsiveness to combination chemotherapy. SCLC is also unusual because of its frequent association with paraneoplastic syndromes, which occur in approximately 20% of patients at presentation.' The most frequent and best understood syndromes are those that result from ectopic hormone secretion by the tumor. Although a large variety of peptide hormones have been shown to be secreted by SCLC cells, ectopic production of antidiuretic hormone (ADH) and corticotropin (adrenocorticotropic hormone; ACTH) are seen most frequently.' The production of such hormones may or may not be associated with overt clinical syndromes, depending on the functional activity and level of the ectopic hormone produced. Clinical Cushing's syndrome secondary to ectopic ACTH production is uncommon, occurring in approximately 5% of all SCLC patients.' However, biochemical evidence of hypercortisolism can be detected in up to 50% of patients,2'3 and it is possible to identify immunoreactive ACTH in almost all SCLC tissue extracts.4 The relationship between ectopic ACTH production and tumor burden has been difficult to study because of the small number of patients with clinical evidence of
proximal myopathy (61%). All patients had elevated plasma and urinary free cortisol levels; 22 had a hypokalemic alkalosis, and 13 had hyperglycemia. Only one patient had a normal ACTH level. The response rate (complete plus partial) to chemotherapy for patients who had the syndrome diagnosed at initial presentation of SCLC was only 46%, and their median survival was only 3.57 months. Only two patients achieved complete normalization of all hormone parameters, and neither experienced hormone relapse at the time of SCLC relapse. Complications of therapy included gastrointestinal (GI) ulceration (six patients), GI bleeding (four), perforation of a duodenal ulcer (one), pneumonia (10), septic shock (three), and fungal infections (five). Conclusion: Ectopic ACTH production is associated with a low response to chemotherapy, short survival, and a high rate of complication to therapy. J Clin Oncol 10:21-27. © 1992 by American Society of Clinical Oncology. the syndrome. One point of apparent consensus is that patterns of ACTH secretion vary considerably. Importantly, plasma ACTH levels do not necessarily correlate directly with tumor burden, and relapse of a hormonesecreting tumor after response is not always accompa-
nied by resumption of hormone secretion. As a result, 56 ACTH can not be used as a tumor marker. , The impact of ectopic ACTH secretion associated
with clinical Cushing's syndrome on disease course and overall survival is unclear. Earlier studies suggested that
ectopic ACTH production might be associated with 37 8 more extensive disease and a worse prognosis, " ' but the
total number of patients reported in these studies is
small. Furthermore, the impact of aggressive combination chemotherapy and new and more effective means of
treating the metabolic abnormalities caused by hypercortisolism have not been assessed. In an attempt to resolve some of these issues, we
From the Division of Hematology and Oncology, Department of Medicine, Toronto Hospital, Toronto General Division, Toronto; and the University of Toronto, Toronto, Canada. Submitted May 22, 1991; acceptedJuly 26, 1991. Address reprintrequests to FrancesA. Shepherd, MD, M/L W2-035, Toronto General Hospital, 200 Elizabeth St, Toronto, Ontario M5G 2C4, Canada. © 1992 by American Society of ClinicalOncology. 0732-183X/9211001-0022$3.00/0
Journalof Clinical Oncology, Vol 10, No 1 (January), 1992: pp 21-27
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
21
22
SHEPHERD ET AL
reviewed the records of all patients with SCLC seen at our institution over a 10-year period to identify patients with ectopic ACTH production associated with clinical Cushing's syndrome. We report here the results of this review. PATIENTS AND METHODS We undertook a 10-year (1980 to 1990) retrospective chart review of 545 patients with histologically or cytologically confirmed SCLC treated at the Toronto General Hospital. From this review, 23 patients with ectopic ACTH production were identified. Patients were considered to have an ectopic ACTH syndrome when there were clinical signs and symptoms of excess corticosteroid production together with at least two of the following conditions: hypokalemia defined as a serum potassium level less than 3.0 mmol/L, elevated plasma ACTH (> 22 pmol/L), elevated plasma cortisol (> 660 nmol/L) with loss of diurnal variation and/or lack of suppressibility by dexamethasone, or an elevated 24-hour urinary free cortisol level (> 400 nmol/d). Pretreatment clinical staging was performed for all patients and included a complete history and physical examination, determination of Eastern Cooperative Oncology Group (ECOG) performance status, a complete blood cell count, and a biochemical profile. Radiologic staging procedures included a chest x-ray, a radionuclide bone scan, an ultrasound or computed tomographic (CT) scan of the abdomen, either a radionuclide or CT scan of the brain, and a bone marrow aspirate and trephine biopsy. Based on the results of the above investigations, disease was classified as either limited (disease confined to the thorax plus the ipsilateral supraclavicular fossa) or extensive. All patients were treated with combination chemotherapy using either cyclophosphamide, doxorubicin, and vincristine (CAV) or etoposide (VP-16)-cisplatin (VP-CP). CAV consisted of cyclophosphamide 1,000 mg/m2, doxorubicin 50 mg/m2, and vincristine 2 mg all given intravenously at 3-week intervals, usually for six cycles. VP-CP consisted of VP-16 100 mg/m 2 and cisplatin 25 mg/m', both given intravenously daily for 3 days at 3- to 4-week intervals for six cycles. Standard SCLC tumor response criteria were used. The response to ectopic hormone therapy was evaluated by measurement of posttreatment plasma ACTH, serum and urinary free cortisol, potassium, and blood sugar levels. A complete hormone response was defined as normalization of all abnormal biochemical parameters related to the ectopic hormone secretion. Partial response was defined as a 50% reduction in the level of the abnormal biochemical parameter that was above the upper limit of normal. Survival time was measured from the date of diagnosis of SCLC to the date of death. Actuarial survival curves were prepared by means of the Kaplan-Meier method,' and comparisons of survival were performed using a Wilcoxon log-rank sum test.'o
RESULTS We reviewed the records of 545 patients with a diagnosis of SCLC seen at the Toronto General Hospital between 1980 and 1990, and identified 23 patients
(4.5%) who had clinically recognizable hypercortisolism. The characteristics of these 23 patients are listed in Table 1. There were 17 male and six female patients, with a median age of 60 years (range, 43 to 77 years). At the time of initial diagnosis of lung cancer, seven
Table 1. Patient Characteristics
No. of patients Median age, years (range) Male:female ratio Performance status (ECOG) 0, 1 2 3,4
Limited Disease
Extensive Disease
Total
7 59 (47-69) 4:2
16 60 (43-77) 13:4
23 60 (43-77) 17:6
6 0 1
12 1 3
18 1 4
5 2
14 2
19 4
3 4*
10 6
13 10
Paraneoplastic syndrome
Ectopic ACTH only Ectopic ACTH and SIADH Ectopic hormone first noted at
Presentation Relapse Abbreviation: SIADH, syndrome
of inappropriate antidiuretic hor-
mone. *Three patients who, at their initial diagnosis of SCLC, had limited disease had extensive disease at relapse when the ectopic ACTH first appeared.
patients (30%) had limited disease and 16 (70%) had extensive disease. Of the seven patients with limited disease, three had ectopic ACTH production at presentation, and four developed the syndrome at relapse when there was also evidence of hematogenous metastases. Of the 16 patients with extensive disease at presentation, 10 had ectopic ACTH production at presentation, and in six, it developed only at relapse. Two of the patients whose ectopic hormone syndrome was diagnosed at relapse had demonstrated transient hypokalemia earlier in the course of their disease, but the syndrome was only documented at the time of relapse. In total, 13 of the 23 patients had evidence of ectopic ACTH production at the time of their initial diagnosis of SCLC, 10 developed the syndrome at relapse, and 20 had extensive disease when the syndrome was first identified. In the 20 patients who had evidence of hematogenous metastases at the time of onset of ectopic ACTH production, liver metastases were present in 15, bone metastases in 13, brain metastases in seven, and bone marrow involvement was documented in 10 patients (Table 2). In addition, of the three patients who had limited disease and ectopic ACTH at presentation, two had thrombocytopenia and a leukoerythroblastic blood film suggestive of bone marrow involvement, but both had normal bone marrow biopsies. One extensivedisease patient had thrombocytopenia at relapse, but a bone marrow biopsy was not performed to confirm metastatic involvement. In total, 13 patients (56%) probably had bone marrow metastases. All patients had either normal or slightly elevated baseline WBC and absolute granulocyte counts at presentation. The me-
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
23
ECTOPIC ACTH AND SCLC Table 2. Distribution of Metastases
Table 4. Laboratory Findings
Site of Metastases
Frequency (%)
Laboratory Parameter
No. Tested
Liver Bone Bone marrow Confirmed by biopsy Thrombocytopenia and leukoerythroblastic blood films CNS Adrenal gland Other
15 (75) 13(65)
Elevated plasma cortisol level Loss of cortisol, diurnal variation Elevated 24-hour urinary free cortisol Elevated plasma ACTH level Hypokalemia Hyperglycemia Hyponotremia
22 20 14 16 23 22 23
10 (50) 3* 7 (35) 4 (20) 2 (10)
*Two patients had limited disease and normal bone marrow biopsies, and one patient with extensive disease did not have a bone marrow biopsy performed.
dian pretreatment platelet count was 157 x 109/L (range, 29 x 109/L to 430 x 109/L), and eight patients
No. Abnormal (%) 22 20 14 15 22 13 4
(100) (100) (100) (94) (96) (59) (17)
sponse, one achieved partial response, and one patient died early. Only one of the 10 patients with extensive disease achieved complete response, three a partial response, and six had progressive disease. Only eight of the 10 patients received chemotherapy at the time of relapse, and three (30%) achieved partial response.
had counts less than 150 x 109/L.
The clinical features of Cushing's syndrome seen in the 23 patients are listed in Table 3. The most common features included peripheral edema (83%), proximal myopathy (61%), and moon facies (52%). Many of the other physical findings associated with the classical syndrome were also seen, but to a lesser extent. The relevant laboratory findings are listed in Table 4. All patients tested had elevated plasma cortisol levels with loss of diurnal variation, as well as elevated 24-hour urinary free cortisol levels. The plasma ACTH was elevated in 94% of patients, and 96% had hypokalemia associated with a metabolic alkalosis. Hyperglycemia was present in 59% of patients. Four patients had concurrent ectopic ACTH and inappropriate ADH production. Tumor Response The impact of combination chemotherapy on both the tumor burden and ectopic ACTH production is shown in Tables 5 and 6. The overall response rate for patients who had ectopic ACTH syndrome at the time of initial presentation of their SCLC was 46%. One of the three patients with limited disease achieved complete reTable 3. Clinical Features of Cushing's Syndrome Finding Peripheral edema Proximal myopathy Moon facies Buffalo hump Truncal obesity Hyperpigmentation Psychosis Hypertension Ecchymoses Hirsutism Striae
Frequency (%) 19 14 12 8 8 5 5 5 4 1 1
(83) (61) (52) (35) (35) (22) (22) (22) (17) (4) (4)
Hormone Response Of the 13 patients who had ectopic hormone production at the initial presentation of SCLC, only two had a complete response of their ectopic hormone production with normalization of all parameters measured. In neither case did the ectopic hormone syndrome recur at relapse. Seven patients had a partial response of their ectopic ACTH production to antitumor therapy with a reduction of ACTH and serum or urinary cortisol levels, but not to normal. Two patients experienced progressive disease with a progressive increase in their serum cortisol levels in spite of chemotherapy. Two patients were not assessable: one died prematurely due to pneumonia, and in another, postchemotherapy ACTH and cortisol levels were not measured. Of the ten patients who had their ectopic hormone production diagnosed at relapse, there were no complete responses. Five patients achieved a partial hormone response: two after treatment with chemotherapy only, two who were Table 5. Response to Combination Chemotherapy Response by Time of Presentation of Ectopic
Limited
Extensive
ACTH Production
Disease (%)
Disease (%)
Total (%)
3 1 (33) 1 (33) 0 1 (33)
10 1 (10) 3 (30) 6 (60) 0
13 2 (15) 4(31) 6 (46) 1 (8)
At initial presentation of SCLC No. of patients CR PR PD Early death At relapse of SCLC No. of patients No treatment CR PR PD
10 2 (20) 0 3 (30) 5 (50)
Abbreviations: CR, complete response; PR, partial progressive disease.
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
response; PD,
24
SHEPHERD ET AL
Table 6. Summary of Treatment and Response Time of Diagnosis
Stage of SCLC
Response of Tumor
of Ectopic ACTH Production
(at presentation/ at relapse)
to Chemotherapy (primary/relapse)
At initial presentation of SCLC
L
Serum
Urinary Free
PR
PR
NT
NT
N
CR
PR
NT
4
NT
NA PR
NT ? PR
NT NA
NT NT
NT NT
PR
NE
NT
NT
NT
CR
CR
N
NT
NT
PD
PD
NC
T
NT
PD
PR
1
1
NT
PD
PR
PR
CR
N
N
NT
PD
PD
NT
1
NT
PD
PR
N
4
NT
E
PD
PR
4
4
N
E
PR/PR
PR
1
1
NC
L/E
PD/PR
Stable
NC
NT
NC
E
PR/PD
PD
N
NC
T
E
PR/PD
PD
4
T 4
E
CR/PD
PR
NT
NC
4
L/E
CR/PR
PR
NT
4
NT
E
PR/NA
PR
NT
NT
N
L/E
CR/PD
NE
NT
NT
NT
E
PD/NA
PR
NC
N
,
L/E
NA/PD
NE
NT
NT
NT
L E
E
E
E
At relapse of SCLC
Cortisol Cortisol
Response of
Syndrome to Therapy ACTH
$
Comments
K+ N: K+ supplementation continued N: treatment no longer required NT N: treatment no longer required N: never had hypokalemia N: treatment no longer required N: K+ supplementation continued N: K+ supplementation and amiloride continued N: K+ supplementation and spironolactone continued N: K+ supplementation stopped N: K+ supplementation continued N: K+ supplementation and spironolactone continued N: K+ supplementation continued N: K+ supplementation and spironolactone continued N: spironolactone continued N: K+ supplementation and spironolactone continued Fluctuating: K+ supplementation and spironolactone continued N: K+ supplementation and spironolactone N: treatment no longer required N: K+ supplementation continued N: spironolactone continued Fluctuating: K+ supplementation continued N: K+ supplementation continued
Early death
No return of ectopic ACTH at relapse
Treated with ketoconazole Treated with ketoconazole No return of ectopic ACTH at relapse, K+ stopped
Treated with ketoconazole
Treated with ketoconazole No response to ketoconazole or aminoglutethimide Treated with ketoconazole
Treated with ketoconazole
Treated with ketoconazole
Abbreviations: L, limited; E, extensive; NA, not applicable; NT, not tested; NC, no change; N, normal; K+, potassium; T , increased;
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1,,decreased.
25
ECTOPIC ACTH AND SCLC
treated with both chemotherapy and ketoconazole, and one who was treated with ketoconazole alone. One patient had stable hormone levels, two had progressive disease, and two died with progressive malignancy before hormone levels could be reevaluated. Ketoconazole was used in the treatment of eight patients. Only in one patient was ketoconazole given without concurrent chemotherapy, and a partial hormone response was achieved. Two of the remaining seven patients demonstrated no response, and one died before repeat hormone testing was undertaken. Four patients had partial hormone response even though their tumors did not respond clinically to chemotherapy. These patients also had some improvement in their ACTH levels, which suggests that both chemotherapy and ketoconazole may have contributed to the reduction in serum and urinary cortisol levels.
a,
o
-. m
0
Survival
20
10
The overall survival for the 23 patients is shown in Fig 1. The median survival of the whole group was 6.23 months. Within this group, patients with ectopic ACTH at initial presentation of SCLC had a significantly shorter survival (median, 3.57 months) compared with those with ectopic hormone diagnosed at relapse who had a median survival of 10.7 months from the date of diagnosis of their SCLC (P = .0020). This was true both for limited- and extensive-disease patients. The median survival of the limited-disease patients overall was only 27 weeks (range, 1 to 75 weeks), and of extensive-disease patients, 28 weeks (range, 3 to 87 weeks). Limited- and extensive-disease patients who had ectopic hormone secretion diagnosed at their initial presentation of SCLC had median survivals of only 9 and 18 weeks, respectively. Toxicity Significant hematologic toxicity was encountered after chemotherapy. The median nadir granulocyte count was only 0.24 x 109/L (range, 0 to 1.8 x 109/L), and six
patients had nadir granulocyte counts of 0. The median nadir platelet count was 33 x 109/L (range, 10 x 109/L to 306 x 109/L), and five patients had counts less than 20 x 109/L. A high rate of complication was seen during treatment in this series (Table 7). Six patients had gastrointestinal (GI) ulceration associated with upper GI bleeding in four patients and perforation of a duodenal ulcer in one. Serious infection occurred in 10 patients. Ten patients had pneumonia, and three experienced episodes of septic shock. Fungal infections were seen in five patients and included disseminated candidi-
TIME IN MONTHS Fig 1. A comparison of the survival from the date of diagnosis of SCLC for patients with ectopic ACTH diagnosed at presentation (----; n = 13; median survival, 3.57 months) or at relapse (; n = 10; median survival, 10.7 months) (P = .0020). (----) All patients; n = 23; median survival, 6.23 months.
asis in two, superficial candidiasis in two, and pulmonary aspergillosis in one. One patient each suffered massive hemoptysis, hemothorax, and gangrene of the foot. DISCUSSION
Although Cushing's syndrome associated with the ectopic production of ACTH has been reported in association with a variety of malignant tumors, it is seen most frequently in patients with SCLC.' Even in SCLC, the clinical syndrome is rare and is seen in 5% of patients or less. In our review, which was limited to Table 7. Complications During Treatment Complication
No. of Patients
Gastrointestinal ulceration Bleeding Perforation Infection Pneumonia Septic shock Disseminated candidiasis Superficial candidiasis Aspergillosis Massive hemoptysis Bullous impetigo Gangrene of the foot Hemothorax
6 4 1 10 10 3 2 2 1 1 1 1 1
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
26
SHEPHERD ET AL
patients with clinical manifestations of hypercortisolism, 23 of 545 patients (4.5%) were found to have a diagnosis of Cushing's syndrome, an incidence similar to that reported previously. Biochemical evidence of ACTH production is seen much more frequently. When radioimmunoassays are used, ACTH can be detected in virtually 100% of SCLC tissue extracts,4 and elevated plasma levels have ',een reported in one quarter to one third of patients.2 5 Bondy and Gilby3 assessed the endocrine function of 106 untreated patients with SCLC and found that 77% of their patients had abnormalities in at least one test that suggested derangement of adrenocortical function. The median age and male:female ratio of the patients in our study did not differ from that of SCLC overall, and most patients had a good performance status at the time of diagnosis. The most common clinical findings included peripheral edema, proximal myopathy, and moon facies, which were all seen in more than half the patients. Several authors have observed that classical "Cushingoid" features may be absent in patients with SCLC and ectopic ACTH, and that, as in our series, rapidly progressive weight loss, myopathy, edema, and hypertension are often the most prominent features of the syndrome." This may be due to the fact that the patients do not live long enough to develop the classical signs. Hypokalemia, often in association with a mild metabolic alkalosis, was seen in all but one of the patients in our series. In fact, it was usually the finding of a low serum potassium level that suggested the presence of ectopic ACTH secretion and led to subsequent confirmatory tests. Howlett et a11 2 have reported that severe hypokalemia occurs more frequently in ectopic ACTH syndromes than in other causes of Cushing's syndrome. All patients tested had elevated plasma cortisol levels with loss of diurnal variation, and elevated 24-hour urinary free cortisol levels. Only one patient who had markedly elevated serum and urinary cortisols had a normal ACTH level. Normal plasma ACTH levels in the presence of abnormal corticosteroid concentrations have been reported previously by Bondy and Gilby3 and may possibly be explained by the production of a functionally active but incomplete ACTH molecule that is not detected by standard radioimmunoassays. The incidence in our series of either new onset diabetes, or worsening of a previously diagnosed diabetic state was also similar to that seen by Bondy and Gilby, who reported abnormal glucose tolerance tests in 56% of the patients studied in their series. All patients in our series who presented initially with both SCLC and ectopic ACTH syndrome were treated
with combination chemotherapy. Only two patients (one limited and one extensive) achieved complete responses, and four patients (one limited and three extensive) achieved partial responses for an overall response rate of only 46%. The relationship between ectopic ACTH production and tumor behavior is not clearly defined in the literature because of the small number of patients in the reported series to date. Although early series suggested that SCLC associated with Cushing's syndrome might be associated with a worse prognosis, many patients in these series were not treated with the aggressive chemotherapy combinations that are current standard therapy. Our low response rate despite chemotherapy treatment with either CAV or VP-CP suggests that the presence of ectopic ACTH secretion does confer an adverse prognosis and that SCLC associated with ectopic ACTH production may be inherently more resistant to chemotherapy than other SCLC tumors. Cell culture and in vitro sensitivity testing could provide interesting data in this regard. Neither of the two patients who had complete normalization of their hormone levels had a return of their ectopic hormone syndrome at the time of relapse of their SCLC. This phenomenon has been reported previously for both ectopic ACTH and ADH secretion,5 as well as for other 1 paraneoplastic syndromes associated with SCLC." The median survival for the 13 patients with ectopic hormone secretion at presentation was less than 4 months and was significantly lower than that seen in patients who developed their syndrome only at the time of relapse. Abeloff et al8 reported a similar experience. The median survival from the time of diagnosis of SCLC in their series was only 7 months, and no patient survived 1 year. Similarly, in a small series reported by Lokich, 7 three of four patients survived 3 months or less. A closer examination of the cause of death in our series suggests that the reason for such poor survival may be multifactorial. Our review suggests that these patients have a much higher rate of complications during therapy than that seen in patients with SCLC not associated with an ectopic hormone syndrome. Forty-three percent of our patients experienced severe infections, and in all cases, the infection contributed significantly to the cause of death. Furthermore, three patients had invasive fungal infections with pulmonary involvement in all cases. Six patients had GI ulceration, four of these patients had active upper GI bleeding, and one suffered a perforation of a duodenal ulcer. An increased susceptibility to infection, particularly fungal infection, and GI ulceration are all recognized complications of hypercortisolism, whether on a pathologic or iatrogenic basis. However, it interesting that such a high incidence of these
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
27
ECTOPIC ACTH AND SCLC
complications was seen in our patient population, who all had hypercortisolism of fairly brief duration. Abeloff et al' reported pulmonary infection and sepsis in two of five patients in their series, and Lokich' reported pulmonary infiltrates possibly due to Pneumocystis carinii or other fungal organisms in two of four patients. The infectious and hemorrhagic complications after therapy may have resulted, in part, from severe posttreatment neutropenia and thrombocytopenia. More than half of our patients had bone marrow involvement by tumor. Ten were confirmed by biopsy, and three others had significant pretreatment thrombocytopenia and leukoerythroblastic blood films. Only one other review has suggested an association between bone marrow metastases and paraneoplastic syndromes. De la Monte et al"4 identified bone marrow involvement in 76% of 28 patients with paraneoplastic syndromes, which included ectopic ACTH secretion as well as other syndromes associated with SCLC. In a previous review, we identified bone marrow involvement in only 67 of 403 patients with SCLC (16.6%) treated in three consecutive University of Toronto lung cancer study protocols."5 Therefore, we believe that the level of bone marrow involvement seen in this series is unusually high, and we question whether there is perhaps an association between bone marrow involvement and ectopic ACTH secretion. In addition to chemotherapy, eight patients received specific therapy to block adrenocorticosteroid produc-
tion. Eight patients were treated with ketoconazole, and one patient received aminoglutethimide after not responding to ketoconazole. Ketoconazole is an imidazole derivative which blocks corticosteroid production in vitro and in vivo by inhibition of 14-demethylation of cholesterol.'" It has been used successfully in the treatment of ectopic ACTH secretion in association with SCLC'7 as well as other tumors.' Four patients achieved partial remission of their ectopic hormone syndrome despite progression of malignant disease while on chemotherapy. Two patients had progression of their malignant disease associated with worsening of their hormone levels. One patient who did not receive chemotherapy at relapse had a partial response in hormone levels during treatment with ketoconazole alone, and one patient died early before hormone testing could be repeated. We believe that ketoconazole contributed to control of hormone levels in five of the eight patients treated with this medication. In summary, clinical Cushing's syndrome secondary to ectopic ACTH secretion occurs in approximately 5% of patients with SCLC. It is associated with a low response to combination chemotherapy and a high rate of complication during treatment. Although corticosteroid levels may be controlled by the use of corticosteroid blocking agents such as ketoconazole, survival remains poor for this group of patients.
REFERENCES 1. Bunn PA, Ridgway EC: Parancoplastic syndromes, in DeVita V, Hellman S, Rosenberg SA (eds): Cancer Principles and Practice. Philadelphia, PA, Lippincott, 1989. pp 1896-1940 2. Hansen M. Hansen H, Hirsch F, et al: Hormonal polypeptides and amine metabolites in small cell carcinoma of the lung with special reference to stage and subtypes. Cancer 45:1432-1437, 1980 3. Bondy PK, Gilby ED: Endocrine function in small cell undifferentiated carcinoma of the lung. Cancer 50:2147-2153, 1982 4. Gewirtz G, Yalow R: Ectopic ACTH production in carcinoma of the lung. J Clin Invest 53:1022-1032, 1974 5. Gropp C, Havemann K, Scheuer A: Ectopic hormones in lung cancer patients at diagnosis and during therapy. Cancer 46:347354, 1980 6. Hansen M, Hammer M, Hummer L: ACTH, ADH and calcetonin as markers of response and relapse in small-cell carcinoma of the lung. Cancer 46:2062-2067, 1980 7. Lokich JJ: The frequency and clinical biology of the ectopic hormone syndromes of small cell carcinoma. Cancer 50:2111-2114, 1982 8. Abeloff M, Trump D. Baylin S: Ectopic adrenocorticotrophic (ACTH) syndrome and small cell carcinoma of the lung-Assessment of clinical implications in patients on combination chemotherapy. Cancer 48:1082-1087, 1981 9. Kaplan EL, Meier P: Non-parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
10. Gehan EA: A generalized Wilcoxon test for comparing arbitrarily. Singly censored samples. Biometrika 52:203-223. 1982 11. Howlett TA, Rees LH, Besser GM: Cushing's syndrome. Clin Endocrinol Metabol 14:911-945, 1985 12. Howlett TA, Drury PL, Doniach PI, et al: Diagnosis and management of ACTH-dependent Cushing's syndrome: Comparison of the features in ectopic and pituitary ACTH production. Clin Endocrinol 24:699-713, 1986 13. Clamon G, Evans W, Shepherd F, et al: Myasthenic syndrome and small cell cancer of lung. Variable response to antineoplastic therapy. Arch Intern Med 144:99-100, 1984 14. De la Monte S, Hutchins G, Moore G: Paraneoplastic syndromes and constitutional symptoms in prediction of metastatic behavior of small cell carcinoma of the lung. Am J Med 77:851-856, 1984 15. Campling B, Ouirt I, DeBoer G, et al: Is bone marrow examination in small cell lung cancer really necessary? Ann Intern Med 105:508-512, 1986 16. Pont A, Williams P, Ltnose D, et al: Ketoconazole blocks adrenal steroid synthesis. Ann Intern Med 97:370-372, 1982 17. Shepherd F, Hoffert B, Evans W, et al: Ketoconazole: Use in the treatment of ectopic adrenocorticotropic hormone production and Cushing's syndrome in small-cell lung cance . Arch Intern Med 145:863-864, 1985 18. Soniro N: The use of ketoconazole as an inhibitor of steroid production. N Engl J Med 317:812-818. 1987
Downloaded from ascopubs.org by UNIVERSITY of OTAGO on April 21, 2019 from 139.080.135.089 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
