The chemotherapy of metastatic gastric adenocarcinomas with hypersecretion of {alpha}-fetoprotein or ß-human chorionic gonadotrophin: report of two cases

N. Germann1, M. Gross-Goupil1, E. Wasserman1, J.-F. Emile2, J.-L. Misset1, M. Reynes2 and F. Goldwasser1,+

1Service de Cancérologie and 2Laboratoire d’Anatomopathologie, Hôpital Paul Brousse, Villejuif, France

Received 9 January 2001; revised and accepted 11 June 2001.

Abstract

The chemotherapy of advanced gastric adenocarcinomas (GAs) is based on agents such as cisplatin, 5-fluorouracil and anthracyclins. Reproducible objective response rates are reported as ~40%. However, the median survival remains short, not exceeding 10 months. Amongst GA, a subset of tumours with increased plasma {alpha}-fetoprotein ({alpha}FP) and/or ß human chorionic gonadotrophin (ßHCG) levels form a well-defined histopathological entity. This subgroup has been associated with poor prognosis, due to the presence of poorly differentiated and rapidly proliferating cells. No specific chemotherapy has been proposed for this particular form of GA. We report two cases of patients with GA and hypersecretion of {alpha}FP and/or ßHCG. Despite bulky liver metastases and resistance to two standard chemotherapy regimens, both patients exhibited sensitivity to chemotherapy combining bleomycin, oxaliplatin and etoposide. These results suggest that patients with this particular subset of GA may benefit from chemotherapy regimens similar to those given to germ-cell tumour patients.

Key words: adenocarcinoma, {alpha}-fetoprotein, ß human chorionic gonadotrophin, chemotherapy, gastric

Introduction

A rare subset of tumours with hyperexcretion of tumour embryonic markers has been identified amongst gastric adenocarcinomas (GAs) [13]. ß human chorionic gonadotrophin (ßHCG) is an embryonic protein secreted in the syncitiotrophoblast of the placenta [4]. {alpha}-fetoprotein ({alpha}FP) is an embryonic protein secreted by the yolksac, by the fetal liver and in the gut [5]. The determination of plasma ßHCG and {alpha}FP levels is routinely used for the diagnosis and follow-up of patients with germ-cell tumours, while {alpha}FP is also a marker for hepatoblastoma and hepatocellular carcinomas. Many adenocarcinomas have been described as being rarely associated with ßHCG hypersecretion including lung, colon, cervix and endometrial cancers [1]. Amongst these adenocarcinomas, the gastric origin was one of the most frequent. The incidence of this subset of GA ranges from 10.9 to 16.7% for ßHCG and from 1.3 to 19% for {alpha}FP [1, 2]. Tumours with {alpha}FP or ßHCG secretion are associated with a high proliferation rate and poor prognosis [3, 4]. Unfortunately, no information is available on their sensitivity to standard GA chemotherapy or to other therapeutic approaches. We report the outcome of two patients with a GA secreting ßHCG and/or {alpha}FP. Both patients were first treated with standard GA chemotherapy, including 5-fluorouracil (5-FU), cisplatin and either adriamycin or 4'epi-adriamycin. Their best response to chemotherapy, although heavily pretreated and with bulky disease, was observed using a third-line regimen combining bleomycin, oxaliplatin and etoposide (BOXE), and targeting the ßHCG- and/or {alpha}FP-secreting tumour subpopulation. These two case reports suggest that GAs with a rapidly growing undifferentiated contingent, secreting {alpha}FP and/or ßHCG may benefit from chemotherapy agents active in germ-cell tumours.

Case report 1

In March 1996, epigastric pains revealed a gastric linitis in a 40-year-old man. There was cryptorchism in his history. A partial oesogastrectomy associated with a splenectomy and a gastroplasty was performed. Histology of the gastric tumour revealed a weakly mucosecretant adenocarcinoma with contingent of independent cells. Because of nodal involvement, the patient received post-surgical chemotherapy with 5-FU, methotrexate and adriamycin (FAMTX protocol). He developed a liver metastasis while on therapy. Another chemotherapy regimen combining 5-FU, epirubicin and cisplatin (ECF protocol) was initiated. Stable disease was the best tumour response after eight cycles and a surgical resection of the unique liver metastasis was performed. The histology of the liver metastasis confirmed the presence of a weakly mucosecretant adenocarcinoma with trabecular architecture, different from the primitive tumour with some contingents compatible with the diagnosis of embryonic carcinoma. Immunohistochemical staining was negative for ßHCG and {alpha}FP in the liver metastasis. Post-surgical treatment with cimetidin was given. In July 1997, 3 months after the last ECF cycle, liver disease progression was diagnosed and ECF protocol was reintroduced. In October 1997, after three more cycles of ECF, an abdominal computer tomography (CT) scan re-vealed a major hepatic tumour progression. The clinical status became altered (with a performance status of 2), the patient was subicteric, had major hepatomegaly with severe liver pain and biological evidence of metastatic liver failure [bilirubin 35 mmol/l, 2 x upper normal value (N)]. His life expectancy was evaluated as less than 1 month. The plasma ßHCG and {alpha}FP levels were high at 111 IU/l (22 N) and 1400 ng/ml (140 N), respectively. The carcinoembryonic antigen (CEA) level was 50 ng/ml (10 N) and the CA 19-9 level was 3609 IU/ml (100 N). Ultrasonography of the testicles revealed a left hydrocele. There was no retroperitoneal lymphadenopathy on the abdominal CT scan. Chemotherapy targeting the ßHCG- and {alpha}FP-secreting population was initiated using a BOXE regimen. The general condition improved after the first cycle, abdominal pains disappeared, plasma ßHCG level returned to normal and the plasma {alpha}FP level decreased by 80% within 3 weeks, with an apparent half-life of less than 1 week. In December 1997, after three cycles, the performance status was normal. The {alpha}FP level had fallen by 96%, the CEA level decreased by 50% and the CA 19-9 level by 90%. Haematological toxicity was evaluated weekly using the NCI-CTC common criteria. The patient presented a febrile grade 3 neutropenia with grade 4 thrombocytopenia after the first cycle. While his general condition improved, the subsequent cycles were better tolerated with grade 3 afebrile neutropenia and grade 3 thrombocytopenia. In February 1998, an abdominal CT scan, performed after six cycles, showed a partial response of more than 75% of the liver metastases (Figure 1). However, hepatic pain reappeared in March 1998 with a concomitant rise in tumour markers. Despite two other chemotherapy regimens based on ifosfamide or irinotecan hydrochloride (CPT-11), the liver metastases progressed, the patient developed liver failure and died in August 1998, 29 months after diagnosis and 10 months after having started the BOXE regimen for rapidly progressing liver metastases. At necropsy, a massive liver involvement with large necrosis was found. No embryonic carcinoma contingent was found. The testicles, the retroperitoneum and the mediastinum were macroscopically normal.



View larger version (67K):
[in this window]
[in a new window]
 
Figure 1. Liver metastases before the first cycle of BOXE (left) and following 6 cycles (right).

 
Case report 2

A 75-year-old man presented in September 1997 with epigastric pain, hiccups, dysphagia and alteration of his general health. He had no particular history of gastric disorders. The gastric fibroscopy revealed a tumour of the cardia. A thoracic and abdominal CT scan showed liver and lung metastases. Histology of the gastric tumour revealed a poorly differentiated adenocarcinoma, weakly mucosecreting and without an embryonic contingent. Immunohistochemical investigations were negative for both ßHCG and {alpha}FP. At diagnosis, the plasma {alpha}FP level was elevated to 40 ng/ml (7 N), the plasma CA 19-9 level was 85 IU/ml (2 N), plasma ßHCG and CEA levels were normal. The ultrasound examination of the testicles was normal. No retroperitoneal lymphadenopathies were detectable on the abdominal CT scan. Chemotherapy combining theprubicin (pirarubicin or THP-adriamycin), mitomycin C, folinic acid, 5-FU and cisplatin was initiated. After one cycle, a functional improvement was noted, dysphagia was less severe and the hiccups had disappeared. But after the second cycle, the clinical symptoms worsened and a CT scan and gastroscopy documented a disease progression. The plasma {alpha}FP level rose to 350 ng/ml and the plasma CA 19-9 level reached 95 IU/ml. In November 1997, a second-line investigational therapy combining CPT-11 and oxaliplatin resulted in a partial response of the lung metastases but this was associated with hepatic progression. The patient rapidly developed hepatomegaly with deformation of the abdomen. The plasma {alpha}FP level was still rising at 4700 ng/ml (700 N) and the plasma CA 19-9 level rose to 170 IU/ml (5 N). A third-line chemotherapy targeting the {alpha}FP-secreting tumour cells was initiated using a BOXE regimen, in March 1998. The plasma {alpha}FP levels fell by >50% within the first 8 days. The patient had a fatal septic shock during aplasia (grade 4 febrile neutropenia, grade 4 thrombocytopenia) after this first cycle. The necropsy revealed important tumour necrosis both in the cardia and in the liver. An anatomopathological complete remission of the lung metastases was noted. The immunohistochemical staining was positive for {alpha}FP in the gastric tumour. The testicles, the retroperitoneum and the mediastinum were macroscopically normal.

Discussion

These two clinical cases show that GA patients with hypersecretion of {alpha}FP or ßHCG may be resistant to standard GA chemotherapy but sensitive to a third-line treatment, similar to the bleomycin, etoposide, cisplatin (BEP) regimen used in germ-cell tumours. Hence, in patients who exhibit a similar history of rapid, and even acute, clinical tumour progression (the patients could indicate the precise day of disease acceleration due to onset of a sudden liver pain), massive liver involvement with large zones of necrosis and low differentiation of the tumour, it may be appropriate to measure plasma ßHCG and {alpha}FP levels either from diagnosis (patient 2) or following a period of indolent disease (patient 1).

The gastric origin of the hypersecretion of ßHCG and {alpha}FP in this type of disease has been demonstrated by immunohistochemical investigations [7, 8]. However, these methods may not be sensitive enough to detect secreting cells in the presence of large zones of tumour necrosis, as observed in the liver metastases of our two patients. We could only detect {alpha}FP in the primary gastric tumour of one of them.

A study compared 27 samples of GA tumours with hypersecretion of {alpha}FP, with 505 {alpha}FP-negative samples [9]. Despite no anatomopathological difference between the two groups regarding the gastric tumour type, {alpha}FP-secreting tumours were more frequently associated with liver metastases (72% compared with 9.7%, P <0.01), lymph node and venous invasion. It was unclear whether {alpha}FP secretion reflected the aggressiveness or advanced stage of the tumour. The 5-year survival of {alpha}FP-secreting GA was 11.6% compared with 52.8% in non-{alpha}FP-secreting GA (P <0.001). In another series of 54 patients with early GA, {alpha}FP correlated with low differentiation (P <0.034) and poor prognosis (P <0.0001) [3]. {alpha}FP-producing GA had high proliferation activity, low apoptosis and rich neovascularisation. Hence, these groups of GA appeared more aggressive and more rapidly progressive than common GA [10].

Bone metastases are five times more frequent in ßHCG-secreting GA than in other GA (P <0.005) [11]. ßHCG also correlated with a worse prognosis in a multivariate analysis of 290 GA patients (hazard ratio =1.7, 95% confidence interval 1.1–2.8) [6].

The most active anticancer agents in GA patients in phase II studies achieving objective tumour response rates of at least 15% are 5-FU, anthracyclins, mitomycin-C, cisplatin [12], etoposide [13] and, more recently, CPT-11 [14] and docetaxel [15]. The response rate to combination chemotherapy may exceed 50% [12]. However, the median survival of metastatic GA patients remains short, not exceeding 10 months [12]. After failure of first-line chemotherapy, no treatment has achieved a significant tumour response rate. Both patients had received two previous chemotherapy regimens and were resistant or refractory to 5-FU, anthracyclins and cisplatin (Table 1). In the first clinical case, a >75% partial response could be obtained on bulky liver metastases with BOXE (Figure 1). This dramatic effect was associated with major clinical benefit and survival gain. In both cases, we noted a fall in plasma {alpha}FP level with an apparent plasma half-life <7 days. Evidence of histological tumour response was also obtained at necropsy for both patients.


View this table:
[in this window]
[in a new window]
 
Table 1. Main features of the two advanced gastric cancer patients
 
The aggressiveness and the secretion of {alpha}FP and ßHCG invited the use of a treatment similar to chemotherapy protocols given in germ-cell tumours. The BOXE chemotherapy regimen is shown in Table 2. Oxaliplatin was given prior to etoposide. Etoposide infusion was repeated for 3 days. Bleomycin was given over 2 days as a 10 mg i.v. bolus, followed by a 10 mg i.v. continuous infusion over 24 h. We used a continuous infusion of bleomycin based on a previous demonstration that this mode of administration is at least as active as i.v. bolus and is less toxic [16]. This regimen was repeated every 3 weeks. The response rate of etoposide as a first-line treatment in GA patients is 21% [13], but etoposide was not found to be active as a second-line therapy [17]. Etoposide is a DNA topoisomerase II poison [18]. DNA topoisomerase II{alpha} expression correlates with high proliferation [19, 20]. As a result, etoposide is a major component of combination chemotherapy in germ-cell tumours, small-cell lung cancers, and aggressive non-Hodgkin’s lymphomas. Hence, etoposide might be more active in GA with a high proliferation rate and low differentiation such as GA with hypersecretion of embryonic markers. Bleomycin is not commonly used in GA but is also a key drug in malignancies with high proliferation rates, such as non-Hodgkin’s lymphomas [21]. Moreover, bleomycin is used as a standard treatment for germ-cell tumours. Oxaliplatin is a diaminocyclohexane (DACH) platinum derivative with in vitro and clinical activity in cisplatin-resistant tumours and mismatch-repair deficient cells [22]. We previously observed, during a phase I clinical trial, that advanced gastric cancer patients could be sensitive to the combination of CPT-11 and oxaliplatin [23].


View this table:
[in this window]
[in a new window]
 
Table 2. The BOXE regimen
 
Despite spectacular progress in our knowledge of oncogenesis, medical oncologists can not routinely tailor the chemotherapy of individual patients on the basis of molecular characteristics of their tumour. Nevertheless, correlations are emerging between some phenotypic abnormalities frequently found in cancer cells and antitumour activity. Here, the rapid evolution and the presence of an undifferentiated tumour led us to choose cytotoxic agents other than those commonly used in GA patients. The dramatic change in growth rate was noteworthy. The modifications of the tumour phenotype may explain both the resistance to standard treatment and the sensitivity to agents known to be active in undifferentiated and rapidly growing tumours. However, in most gastric cancer patients, our chemotherapy regimen is unlikely to be as active as reported in the first case report. We do not recommend this regimen as a second-line therapy for advanced cancer patients. Nevertheless, when the same clinical situation is met, we propose to measure {alpha}FP and ßHCG. If there is evidence of hypersecretion of these proteins, we suggest specific targeting of this population of cells with a chemotherapy, such as BOXE.

Footnotes

+ Correspondence to: Dr F. Goldwasser, Service de cancérologie, Hôpital Paul Brousse, 14 Ave Paul Vaillant Couturier, 94800 Villejuif, France. Tel: +33-145-59-36-49; Fax: +33-140-96-99-47; E-mail: f.goldwasser@wanadoo.fr Back

References

1. Blackman MR, Weintraub BD, Rosen SW et al. Human placental and pituitary glycoprotein hormones and their subunits as tumor markers: a quantitative assessment. J Natl Cancer Inst 1980; 65: 81–93.[ISI][Medline]

2. Takahashi Y, Mai M, Ogino T et al. Clinicopathological study of AFP producing gastric cancer—significance of AFP in gastric cancer. Nippon Geka Gakkai Zasshi 1987; 88: 696–700.[Medline]

3. Ishikura H, Fukasawa Y, Ogasawara K et al. An AFP-producing gastric carcinoma with features of hepatic differentiation. A case report. Cancer 1985; 56: 840–848.[ISI][Medline]

4. Dreskin RB, Spicer SS, Greene WB. Ultrastructural localization of chorionic gonadotrophin in human term placenta. J Histochem Cytochem 1970; 18: 862–874.[ISI][Medline]

5. Gitlin D, Perricelli A, Gitlin GM. Synthesis of alpha-fetoprotein by liver, yolk sac and gastrointestinal tract of the human conceptus. Cancer Res 1972; 32: 979–982.[ISI][Medline]

6. Webb A, Scott-Mackie P, Cunningham D et al. The prognostic value of serum and immunohistochemical tumour markers in advanced gastric cancer. Eur J Cancer 1996; 32A: 63–68.

7. Manabe T, Adachi M, Hirao K. Human chorionic gonadotrophin in normal, inflammatory, and carcinomatous gastric tissue. Gastroenterology 1985; 89: 1319–1325.[ISI][Medline]

8. Kato T, Takahashi H, Ida Y et al. Immunocytohistochemical studies of AFP producing gastric cancer—cytomorphology and characteristics of AFP-positive cells. Rinsho Byori 1993; 41: 1024–1030.[Medline]

9. Chang YC, Nagasue N, Abe S et al. Comparison between the clinicopathologic features of AFP-positive and AFP-negative gastric cancers. Am J Gastroenterol 1992; 87: 321–325.[ISI][Medline]

10. Koide N, Nishio A, Igarashi J et al. Alpha-fetoprotein-producing gastric cancer: histochemical analysis of cell proliferation, apoptosis, and angiogenesis. Am J Gastroenterol 1999; 94: 1658–1663.[ISI][Medline]

11. Yonemura Y, Oyama S, Sugiyama K et al. Human chorionic gonadotrophin in gastric carcinoma. A useful marker for bone metastasis. Int Surg 1989; 74: 84–87.[ISI][Medline]

12. Alexander HR, Kelsen DG, Tepper JC. Cancer of the stomach. In DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology. Philadelphia, PA: Lippincott-Raven 1997; 1: 1021–1054.

13. Macdonald JS, Havlin KA. Etoposide in gastric cancer. Semin Oncol 1992; 19 (6 Suppl 13): 59–62.[Medline]

14. Futatsuki K, Wakui A, Nakao I et al. Late phase II study of irinotecan hydrochloride (CPT-11) in advanced gastric cancer. CPT-11 Gastrointestinal Cancer Study Group. Gan To Kagaku Ryoho 1994; 21: 1033–1038.[Medline]

15. Sulkes A, Smyth J, Sessa C et al. Docetaxel (Taxotere) in advanced gastric cancer: results of a phase II clinical trial. EORTC Early Clinical Trials Group. Br J Cancer 1994; 70: 380–383.[ISI][Medline]

16. Krakoff IH, Cvitkovic E, Currie V et al. Clinical pharmacologic and therapeutic studies of bleomycin given by continuous infusion. Cancer 1977; 40: 2027–2037.[ISI][Medline]

17. Yamao T, Shimada Y, Ohtsu A et al. An early phase II study of etoposide (VP-16) in advanced gastric cancer. Jpn J Clin Oncol 1996; 26: 36–41.[Abstract]

18. Pommier Y, Fesen M, Goldwasser F. DNA topoisomerase II in- hibitors: Epipodophyllotoxins, m-Amsa and Ellipticines derivatives. In Chabner BA, Collins JM (eds): Cancer Chemotherapy and Biotherapy: Principles and Practice. Philadelphia, PA: JB Lippincott Company 1996; 435–461.

19. Jarvinen TA, Kononen J, Pelto-Huikko M, Isola J. Expression of topoisomerase IIalpha is associated with rapid cell proliferation, aneuploidy, and c-erbB2 overexpression in breast cancer. Am J Pathol 1996; 148: 2073–2082.[Abstract]

20. D’Andrea MR, Farber PA, Foglesong PD. Immunohistochemical detection of DNA topoisomerase II{alpha} and IIß compared with detection of Ki-67, a marker of cellular proliferation, in human tumours. Appl Immunohistochem 1994; 2: 177–185.[ISI]

21. Loehrer PJ, Johnson DH, Elson P et al. Importance of bleomycin in favorable-prognosis disseminated germ cell tumours: an Eastern Cooperative Oncology Group trial. J Clin Oncol 1995; 13: 470–476.[Abstract]

22. Raymond E, Chaney SG, Taamma A, Cvitkovic E. Oxaliplatin: a review of preclinical and clinical studies. Ann Oncol 1998; 9: 1053–1071.[Abstract]

23. Goldwasser F, Gross-Goupil M, Tigaud JM et al. Dose escalation of CPT-11 in combination with oxaliplatin using an every two weeks schedule: a phase I study in advanced gastrointestinal cancer patients. Ann Oncol 2000; 11: 1463–1470.[Abstract]





This Article
Abstract
Full Text (PDF)
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (1)
Disclaimer
Request Permissions
Google Scholar
Articles by Germann, N.
Articles by Goldwasser, F.
PubMed
PubMed Citation
Articles by Germann, N.
Articles by Goldwasser, F.