Somatostatin analogs and prednisone in advanced refractory thymic tumors

Abstract

BACKGROUND

Therapeutic options to cure advanced, recurrent, and metastatic thymic tumors are limited. Evidence of a high uptake of indium-labeled octreotide (¹¹¹In-DTPA-D-Phe¹-octreotide) in thymic tumors and the curative application of somatostatin analogs and prednisone in one patient with thymoma and pure red cell aplasia led the authors to start a Phase II study.

METHODS

Sixteen patients with advanced thymic tumors, unresponsive to conventional chemotherapeutic regimens, were enrolled in the study. The schedule includes administration of somatostatin analog octreotide (1.5 mg/day subcutaneously) associated with prednisone (0.6 mg/kg/day orally for 3 months, 0.2 mg/kg/day orally during follow-up). In 8 cases, octreotide was replaced by the long-acting analog lanreotide (30 mg/every 14 days intramuscolarly). Treatment was prolonged until progression of disease was documented. Overall response rate, survival, progression free survival, and toxicity were evaluated.

RESULTS

The overall response rate among 16 evaluable patients was 37%. One patient (6%) had a complete response, 5 (31%) had a partial response, 6 obtained a stabilization of disease, and 4 progressed during the treatment. After a median follow-up of 43 months, the median survival was 15 months, and median time to progression was 14 months. Treatment was generally well tolerated with acceptable toxicity: cholelithiasis (1 patient), Grade 2 cushingoid appearance (3 patients), Grade 1 diarrhea (5 patients), Grade 2 hyperglycemia (3 patients).

CONCLUSIONS

Treatment with somatostatin analogs and prednisone has shown efficacy in patients with recurrent and metastatic malignant thymic tumors refractory to standard therapeutic options. The results obtained are very satisfactory given the lack of effective alternative treatments. Such therapy is not burdened by the same toxicity of chemotherapy; thus, it can be administered to heavily pretreated patients. Somatostatin analogs and prednisone are well tolerated, and the long-acting analog lanreotide, which requires fewer injections, improves patients' compliance. Cancer 2002;94:1414–20. © 2002 American Cancer Society.

DOI 10.1002/cncr.10374

Thymic tumors are rather rare epithelial tumors developing within the anterior mediastinum and generally exhibiting a slow pattern of growth. Thymic tumors often may be associated with other disorders such as myasthenia gravis, pure red cell aplasia, and hypogammaglobulinemia.1 These immune disorders, whose role in disease outcome is unknown, complicate the management of thymic tumor patients. Malignancy is expressed more often by the infiltration of adjacent organs, which usually does not allow surgery and requires other therapeutic approaches. A multidisciplinary approach based on neoadjuvant chemotherapy, surgical resection, postoperative radiation therapy, and consolidation chemotherapy is promising for unresectable newly diagnosed malignant thymomas.2 Chemotherapy can improve the outcome of invasive or recurrent thymoma.3 The regimens used usually are based on cisplatin. Among them, the most common are PAC (cisplatin, doxorubicin, cyclophosphamide) and ADOC (doxorubicin, cisplatin, vincristine, cyclophosphamide), which achieved overall response rates of 70%4 and 92%,5 respectively, when used in previously untreated patients.

Metastatic thymic tumors are generally chemorefractory to second-, and also third-line chemotherapeutic regimens, and permanent control in heavily pretreated patients is rare.6

New therapeutic strategies are being tried in cases of advanced thymic tumors, i.e., paclitaxel,7, 8 high-dose chemotherapy with peripheral stem cell or with umbilical cord blood transplantation,9 and the somatostatin analog octreotide, alone or associated with prednisone.10

In 1989, Fuller and Verity demonstrated somatostatin gene expression in the rat thymus and suggested that it exerted a possible paracrine role in modulating T-lymphocyte development.11 Reubi et al. identified somatostatin receptors in normal thymus and in thymic carcinoids, but not in thymomas.12, 13 Furthermore, none of the patients submitted to somatostatin receptor scintigraphy had uptake in normal thymus.12

We have demonstrated a high in vivo indium-labeled octreotide (¹¹¹In-DTPA-D-Phe¹-octreotide) uptake in thymic tumors, but not in adult patients with histologically diagnosed benign thymic hyperplasia.14 This prompted us to administer octreotide associated with prednisone in a patient affected by malignant thymoma and pure red cell anemia; the treatment resulted in significant tumor shrinkage and resolution of the anemia, whereas the single drugs used alone were ineffective.15 This suggested that the two compounds exerted a synergistic effect on the thymoma and on the immune mechanisms underlying pure red cell anemia. Of note, normal human thymus expresses somatostatin mRNA and somatostatin receptor subtypes (ssts) sst₁, sst_2A and sst₃, and cultured thymic epithelial cells expressing sst₁ and sst_2A mRNA do not proliferate under somatostatin treatment.16

These data provided the rationale for a Phase II study to evaluate the efficacy and toxicity of treatment with somatostatin analogs and prednisone in patients with thymic tumors refractory to chemotherapy. Given the similar affinity profile of octreotide and lanreotide,17 the comparable effectiveness of lanreotide and regular somatostatin analogs in the treatment of gastrointestinal neuroendocrine tumors,18 and the advantages of long-acting lanreotide over multiple daily injections of octreotide in terms of patient compliance and costs, 8 patients of 16 were treated with the long-acting somatostatin analog lanreotide, which was the only available long-acting formulation at the time of study beginning.

PATIENTS AND METHODS

Since 1992, we have evaluated the role of somatostatin analogs to image and to treat patients with thymic tumors.

Selection of Patients

Sixteen patients (8 males, 8 females; mean age, 50.5 years; range, 23–74) with histologically confirmed thymic tumors were enrolled in the study. All patients gave informed consent to the trial. Surgical resection was performed in nine patients, whereas seven inoperable patients underwent diagnostic biopsies only. The patients' profiles at study entry are reported in Table 1. The eligibility criteria were 1) presence of histologically confirmed advanced or recurrent thymic tumors, i.e., epithelial (type A–C thymoma) and the neuroendocrine thymic tumors (as used herein, “thymic tumor” refers collectively to these histologic subtypes); 2) tumors refractory to one chemotherapeutic regimen, at least; 3) positive ¹¹¹In-DTPA-D-Phe¹-octreotide scan; 4) no evidence of gallstones by ultrasound; 5) signed informed consent. Previous radiotherapy and biologic therapy were allowed.

Table 1. Patient Profiles at Study Entry

Patient no.	Gender	Age	Histology	Stage	Sites of disease at presentation	Associated diseases	Surgery	Radiotherapy	Chemotherapy	Other treatments
1	F	56	B2+B3	IVA	Med, pleura	Thy	Yes	CEVC (6 cycles)
T (5 cycles)
2	M	23	SCNC	IVB	Med, lung, scl lymph, ax lymph	B	—	CBDCA + CTX + Pdn (4 cycles) CBDCA + VP-16 (3 cycles) CDDP + STZ (8 cycles) CAVBP/DEP (2 cycles)	Laser endoscopic therapy
3	M	63	SCNC	IVB	Med, bone	PRCA	B	—	CAV (6 cycles)
4	F	48	C	IVA	Med, pleura, lung	Thy	Yes	PEB (6 cycles)
EPI + CTX (2 cycles)
CBDCA + VP-16 (2 cycles)
5	F	67	SCNC	IVA	Med, pleura	B	—	CDDP + CTX (8 cycles)
VP-16 (2 cycles)
6	M	27	B2	II	Med	MG	Thy	—	CDDP + CTX (4 cycles)
7	F	60	C	IVB	Med, lung, liver	B	—	PEB (6 cycles)
8	F	74	B1	IVA	Med, pleura	MG	Thy	Yes	CDDP + CTX (2 cycles)
9	M	72	B2	III	Med, pleura, lung, Barety lymph	Thy	Yes	CHOP (6 cycles) CDDP + VP 16 (6 cycles) CBDCA + EPI + VP 16 (6 cycles)
10	M	48	B3	IVA	Med, pleura	Thy	—	nCT PAC (4 cycles)
ADOC (6 cycles)
11	M	59	B2	IVB	Med, lung, pleura, suprahepatic lesion	B	Yes	VCR + alphaIFN (6 cycles) CBDCA + EPI + CTX (4 cycles) MMC + NVB + VP-16 (4 cycles)	Alpha-2b-IFN rlL-2 (2 MU/twice a week and 1 MU/twice a week) laser endoscopic therapy
12	M	54	B2	IVA	Med, med lymph, pleura, lung	Thy	Yes	nCT CDDP + DOX (4 cycles) MITO (5 cycles)
13	F	30	B3	IVA	Pleura	MG	Thy	Yes	ADOC (3 cycles)
14	F	68	C	III	Med, med lymph	My	B	Yes	PAC (6 cycles)
15	M	70	B1	IVA	Med, pleura, lung	B	—	PEC (2 cycles)
Citofur
16	M	52	B2	IVA	Med, pleura, lung	MG	Thy	Yes	ADOC (6 cycles)

• M: male; F: female; Thy: thymectomy; B: biopsy; SCNC: small cell neuroendocrine carcinoma; med: mediastinum/mediastinal; scl: supraclavicular; ax: axillary; lymph: lymph nodes; PRCA: pure red cell aplasia; MG: myasthenia gravis; My: myeloradiculopathy; nCT: neoadiuvant chemotherapy; CDDP: cisplatin; CTX: cyclophosphamide; EPI: epirubicin; DOX: doxorubicin; Pdn: prednisone; T: paclitaxel; CBDCA: carboplatin; VP16: etoposide; STZ: streptozotocin; MITO: mitoxantrone; MMC: mitomycin C; NVB: vinorelbine; CEVC: cyclophosphamide, epirubicin, vincristine, cisplatin; CAVBP/DEP: cyclophosphamide, doxorubicin, vincristine, bleomycin, cisplatin, prednisone, etoposide; CAV: cyclophosphamide, doxorubicin, vincristine; PEB: cisplatin, etoposide, bleomycin; CHOP: cyclophosphamide, doxorubicin, vincristine, prednisone; PAC: cisplatin, doxorubicin, cyclophosphamide; PEC: cisplatin, epirubicin, cyclophosphamide; ADOC: doxorubicin, cisplatin, vincristine, cyclophosphamide.

Associated Diseases

At diagnosis, four patients (two males and two females) were affected by myasthenia. One patient had pure red cell aplasia. One patient suffered from sensorimotor myeloradiculoneuropathy.

Staging

After surgery, tumors were staged according to Masaoka et al.19 One patient was at Stage II, four patients were at Stage III, seven at Stage IVA, and four at Stage IVB.

Histology

The formalin fixed and paraffin embedded blocks and/or hematoxylin and eosin–stained slides of all tumors, obtained from the archives, were reviewed. According to the World Health Organization histologic classification,20 10 tumors were typed as B1–3 thymomas, 3 as C thymomas (nonorganotypic thymic tumors), and 3 as small cell neuroendocrine carcinomas, i.e., poorly differentiated neuroendocrine carcinomas.

Previous Treatments

Nine patients had received previous radiotherapy. One patient had received immunotherapy (alpha-2b-interferon associated with vincristine and prednisone for 2–3 months and recombinant interleukin-2 at doses of 2 million international units 5 days a week followed by a week of rest for a total of 9 months). All patients previously had received at least first-line chemotherapy. Patient 11 had undergone repeat laser surgery for the removal of an endobronchial polypoid thymoma, which was histologically proven.

Routine Diagnostic Workup and Follow-Up

Patients underwent a monthly clinical examination and routine blood examinations, computed tomography, and/or magnetic resonance every 3 months. Liver ultrasound was performed every 6 months to identify gallbladder disorders, which can occur as a side effect of somatostatin analog treatment. All patients underwent ¹¹¹In-DTPA-D-Phe¹-octreotide scintigraphy (3 mCi [111 MBq]/10 mg of peptide; Mallinckrodt, Patten, the Netherlands) at study entry and every year thereafter according to procedures described elsewhere.14

Treatment Regimen

All patients received the somatostatin analog octreotide (Sandostatina, Novartis Farma, 1.5 mg/day subcutaneously, divided into 3 daily 0.5-mg doses) associated with prednisone (0.6 mg/kg/day orally for 3 months, 0.2 mg/kg/day orally during follow-up). Treatment was stopped when disease progression was documented. In eight cases octreotide was replaced by the long-acting analog lanreotide (Ipstyl, Ipsen Pharmaceuticals, Milan, Italy) 30 mg every 14 days intramuscularly. Concurrent administration of chemotherapy and/or other biologic therapies was not allowed.

Evaluation of Response

Objective tumor response was assessed after 6 months of therapy. Complete response was defined as the disappearance of all known lesions. Partial response was defined as a decrease of 50% or more in the sum of the largest dimensions of measurable lesions, no progression of lesions, and no new lesion. A status of complete or partial response required a confirmatory assessment after a 4-week interval. Stable disease was a decrease of less than 50% or an increase of less than 25% in the sum of the largest dimensions of measurable lesions, and no new lesion. Progressive disease was an increase of 25% or more in the size of one or more measurable lesions, or the appearance of new lesions. The objective response rate was the proportion of complete and partial responses in all enrolled patients. The clinical benefit rate is the ratio between the total number of complete and partial responses and stable diseases lasting more than 6 months and the number of evaluable patients. Time-to-progression and survival curves were constructed according to the Kaplan–Meier method. Toxicity was graded according to the common toxicity criteria of the National Cancer Institute.

RESULTS

The median follow-up was 43 months (range, 1–55 months). As shown in Table 2, there were 1 (6%) complete response, 5 (31%) partial responses, 6 (37%) stable diseases, and 4 instances of disease progression (25%). Thus, the overall response rate (complete plus partial responses) was 37% (95% confidence intervals, 14–60 %). The median survival time was 15 months (95% Brookmeyer-Crowley confidence interval, 12–28), and the median time to progression was 14 months. Estimated curves of overall survival and time to progression are represented, respectively, in Figures 1 and 2. The clinical benefit rate was approximately 70%.

Table 2. Results of Somatostatin Analog-Based Therapy

Patient no.	Response	Survival (mos)	Outcome	Side effects
1	PR	15	D	Cushingoid appearance (Grade 2)
2	PD	12	D	Cushingoid appearance (Grade 2)
3	PR	18	D
4	PD	2	D
5	SD	13	D
6	CR	55	A	Cushingoid appearance (Grade 2)
7	PR	30	D	Hyperglycemia (Grade 2)
Diarrhea (Grade 1)
8	SD	7	D	Hyperglycemia (Grade 2)
Diarrhea (Grade 1)
9	PD	12	D	Cholelithiasis
10	SD	43	A
11	SD	28	D	Diarrhea (Grade 1)
12	PR	31	D
13	PD	1	D
14	SD	22	D	Hyperglycemia (Grade 2)
Diarrhea (Grade 1)
15	SD	23	D
16	PR	14	D	Diarrhea (Grade 1)

• PR: partial response; D: dead. PD: progression of disease; CR: complete response; A: alive; SD: stable disease

Figure 1

Kaplan–Meier estimated curve for overall survival.

Figure 2

Kaplan–Meier estimated curve for time to progression.

The patient obtaining a complete response had the longest survival, 55 months. This was similar to the duration of survival of 47 months in our patient reported previously. Both were affected by B2 thymoma, Stage II and Stage IVB, respectively. Only one was submitted to thymectomy, whereas the other underwent a diagnostic biopsy. Therefore, surgical unresectability does not seem to affect the response to treatment.

Two of the five patients with a partial response had B2 thymoma (Stage III and Stage IVA, respectively). The remaining three patients had Stage IVB to IVC thymoma, Stage IVA combined (B2 plus B3) thymoma, and Stage IVB small cell neuroendocrine carcinoma, respectively. Three of the five underwent surgical resection, whereas the other two underwent biopsy. Survival in patients with partial response ranged from 14 months (Patient 12) to 31 months (Patient 16; median, 18 months).

Of the six patients with stable disease, two were affected by Stage IVA B1 thymoma. The remaining four patients had Stage IVB B2 thymoma, Stage IVA B3 thymoma, Stage IIIC thymoma, and Stage IV small cell neuroendocrine carcinoma, respectively. Survival in patients with stable disease ranged from 7 (Patient 8) to 43 months (Patient 10) for a median duration of 22.5 months. Only two patients of the six underwent thymectomy; the remaining four were considered inoperable. Two patients with stable disease had a long-term survival (28 and 43 months in Patients 11 and 10, respectively). Patient 11 who presented with an advanced B2 thymoma (Stage IVB) received three chemotherapy regimens, immunotherapy and laser endoscopic therapy to remove an endobronchial growth of the thymoma (histologically confirmed) before beginning somatostatin analogs and prednisone. Five patients of six showed signs of mediastinal syndrome, whereas the remaining patient presented myeloradiculopathy.

Patient 2, in whom the disease progressed during therapy, had Stage IVB small cell neuroendocrine carcinoma. He received four chemotherapy regimens before beginning somatostatin analogs plus prednisone. During the latter therapy, he underwent laser endoscopic treatment because of severe respiratory distress. Among the patients with disease progression, three were treated for less than 2 months.

Toxicity was generally acceptable: Grade 1 diarrhea in five patients, Grade 2 cushingoid appearance in three patients, and Grade 2 hyperglycemia in three patients (Table 2). Diarrhea was present only at the beginning of treatment and resolved spontaneously. Treatment was stopped only in one patient (Patient 9), because of the appearance of cholelithiasis, not reversed by ursodiol.

Of the 14 patients who died, 7 suffered from severe infections and 2 had severe myasthenia unresponsive to specific treatment.

DISCUSSION

We report the results obtained with the combination of somatostatin analogs and prednisone in a series of 16 patients with histologically proven invasive, recurrent, or metastatic malignant thymic tumors who were unresponsive to previous conventional therapies. Most patients had been heavily pretreated and were not eligible for further conventional therapies. Consequently, the overall response rate of 37% together with a long-lasting complete response in one patient with B2 thymoma can be considered satisfactory. To our knowledge, the reported results and survival data are unique because they came from advanced chemorefractory patients, considered no more eligible for additional conventional chemotherapy as well as debulking surgery. All enrolled patients had progressive and symptomatic disease at the beginning of therapy. Other published data are related to progressing, but still chemoresponsive patients.4, 21 The few patients with heterogeneous histologic subtypes and stages do not allow for defining any statistical correlations with the response obtained. Nevertheless, neither histology nor disease stage seem to influence the response to treatment. Indeed, treatment response was not limited to tumors that have a predominant or exclusive neuroendocrine phenotype. This feature is characteristic of most types of malignant thymic tumors, including thymic carcinoma.22, 23

Glucocorticoid receptors were identified in both human and murine cultured thymic epithelial cells24 and glucocorticoid hormones modulate thymic epithelial cell proliferation and production of thymic hormones.25 Corticosteroids can induce thymoma regression. Although they act mainly on the lymphocytic component of the tumor, their effectiveness is apparently unrelated to thymoma cell type.26, 27

Somatostatin and somatostatin analogs inhibit the growth of tumor cells. They also inhibit angiogenesis and growth factors (e.g., insulin-like growth factor [IGF]–I).28 In vivo and in vitro data suggest that somatostatin plays an important role in normal and neoplastic thymus, probably by counteracting the local influence of the GH-IGF-I pathway.29, 30

Low doses of dexamethasone enhance the expression of the somatostatin gene in the thymic gland.11 However, given the complex thymic microenvironment, any inference concerning a synergism between somatostatin and steroids can only be speculative.

Somatostatin analogs are routinely used to treat hormone-secreting tumors.28 Their activity also has been tested in other types of tumors in vitro and in vivo.31-34

Somatostatin analogs have been claimed to be moderately effective in treating lymphoproliferative disorders—an observation not supported by prior evaluation with somatostatin receptor scintigraphy.34 In our series, the in vivo demonstration of somatostatin receptors was a primary criterion for treatment with somatostatin analogs.

We previously demonstrated that somatostatin and corticosteroids together, but not as single agents, resolved not only the thymoma, but also pure red cell anemia.15

Somatostatin and steroid interference with immune system functions are known, and they are probably important in the control not only of tumor cells, but also of the thymoma-related immune disorders. The course of the immune alterations and the therapeutic use of somatostatin analogs is difficult to define. Currently, we do not know the exact role of the endocrine and immune actions of the combination of somatostatin analogs and prednisone in thymoma. Nonetheless, this association was effective whereas the use of single agents failed to provide resolution of pure red cell aplasia in both our case report15 and in this series.

In conclusion, the analysis of the results obtained in this series of advanced thymic tumors reveals that this therapeutic approach is promising. Somatostatin analog and prednisone can be considered an effective new treatment with acceptable toxicity for patients unresponsive to standard treatment options. We feel the results of this series certainly merit the development of a larger series, perhaps through a multi-institutional program to fully evaluate this combination in more patients.

Acknowledgements

The authors are indebted to Dr. Robert E. O'Mara, Medical Center, University of Rochester, United States, for advice related to this article.