Peptide receptor radioligand therapy is an effective treatment for the long-term stabilization of malignant gastrinomas

Abstract

BACKGROUND:

Gastrinomas, a rare group of neuroendocrine tumors, are responsible for severe peptic disease and diarrhea. Although symptomatic control may be achieved with proton-pump inhibitors (PPIs) and somatostatin analogues (SSAs), data are limited regarding the possible antitumor effect of the peptide receptor radioligand therapy (PRRT) with radiolabeled SSAs in gastrinoma patients. The goal of this study was to assess the effect of PRRT on symptoms, gastrin secretion, and tumor load in patients with progressive malignant gastrinomas.

METHODS:

We retrospectively studied 11 patients with metastatic gastrinomas followed for a mean period of 6 years. All patients were symptomatically treated with PPIs, and 9 of 11 patients received monthly injections of SSAs; all patients had an Eastern Cooperative Oncology Group score of 0-1, and received PRRT (⁹⁰Yttrium- or ¹⁷⁷Lutetium-DOTATOC) for progressive disease. Serum gastrin measurements and radiological assessment (using the Response Evaluation Criteria in Solid Tumors criteria) were performed before and every 3-6 months following PRRT.

RESULTS:

PRRT induced symptomatic improvement in all patients. The mean serum gastrin decreased significantly from 4831 mI/L to 932.6 mI/L (normal, 40-108 mI/L; P < .001). Periodic radiological surveillance showed complete response in 1 (9%) patient, partial tumor response in 5/11 (45%) patients, and tumor stabilization in 5/11 (45%) patients. In 7/11 (64%) patients, the antitumor effect of PRRT persisted after a median period of 14 months. Four of 11 (36%) patients died due to tumor progression (median time to progression, 11 months); in this group, the mean survival time after the last PRRT was 14 ± 6.9 months.

CONCLUSIONS:

PRRT seems to be a promising tool for the management of patients with inoperable or progressive metastatic gastrinomas. Cancer 2011. © 2010 American Cancer Society.

Gastrinomas, a rare group of neuroendocrine tumors (NETs) that are usually located in the duodenum or pancreas (the “gastrinoma triangle”), hypersecrete gastrin, resulting in Zollinger-Ellison syndrome. The main characteristics of Zollinger-Ellison syndrome are gastric acid hypersecretion (inducing serious peptic ulcers and gastro-esophageal reflux) and diarrhea.1, 2 The syndrome may develop in association with sporadic tumors, which are the most frequent subtype of gastrinomas (≈66%), or as part of the multiple endocrine neoplasia type 1 familial syndrome. Sporadic gastrinomas are reported to be malignant in approximately 40%-85% of cases,3 metastasizing to different sites, including the lymph nodes, liver, and skeleton.

Treatment of gastrinoma is initially a symptomatic approach, through the use of proton pump inhibitors (PPIs). These drugs have changed the primarily acid hypersecretion-associated morbidity to one of tumor growth and spread.4 Following symptom control, surgical approach is advocated for the tumor excision before metastatic spread occurs.

The presence of peripancreatic lymph node metastases may still permit cure if the involved nodes can be completely resected together with the primary tumor5-7; however, an aggressive approach to liver metastases is controversial.6, 7 The primary determinants of survival in gastrinoma are size of the primary tumor and liver metastases; age at diagnosis, sex, presence of lymph node metastases, association with multiple endocrine neoplasia type 1, and method of ulcer treatment were all reported to have no effect.8 The tumor-related mortality in patients with metastatic gastrinoma was reported to be as high as 79%,9 with at least one-half of patients dying from gastrinoma-related causes and none dying from acid-related causes. Except for large primary pancreatic tumors and the extent of liver metastases, the presence of bone metastases or paraneoplastic syndromes (eg, Cushing syndrome) were also reported to affect patient survival.9

In patients with metastatic disease, in whom it is unclear whether surgical intervention offers any benefit on survival or on time to tumor progression (TTP), less aggressive approaches (including somatostatin analogues, peptide receptor radioligand therapy (PRRT), interferon, chemotherapy, and chemo-embolization) have been suggested.10 However, the role of adjuvant therapies specifically for gastrinoma patients remains unclear.

We conducted a retrospective study to define clinical, biochemical, and radiological response, and time to tumor progression in a group of patients with metastatic gastrinomas treated with PRRT (⁹⁰Yttrium [⁹⁰Y]-/¹⁷⁷Lutetium [¹⁷⁷Lu]-DOTATOC or [¹⁷⁷Lu]-DOTATATE) for progressive disease.

MATERIALS AND METHODS

Eleven consecutive patients with metastatic gastrinomas were selected from a database including all patients with NETs treated at 2 referral centers in Israel; patients files have been retrospectively studied for a mean period of 6 years (between 2004 and 2010). The cases included patients diagnosed, treated, and followed in the NET clinics, and information on clinical presentation, biochemical profile, radiological assessment, histopathological findings, and extent of the disease (using the World Health Organization classification)11 were recorded. In addition, the application of other therapeutic modalities including surgery, medical (eg, somatostatin analogues, transarterial chemoembolization) or radioactive therapeutic modalities, and long-term outcome of these tumors were recorded. Patients were assessed clinically, biochemically, and radiologically before PRRT and thereafter at 3- to 6-month intervals according to a uniform protocol described below.

The study included patients with sporadic tumors as well as tumors being diagnosed as part of the familial multiple endocrine neoplasia type 1 syndrome.

All patients were treated with PPIs for acid-related symptoms, and 9 of 11 patients also received monthly injections of somatostatin analogues (SSAs) as antiproliferative treatment: octreotide LAR (30 mg/mo; Novartis, Basel, Switzerland; n = 8) or lanreotide Autogel (120 mg/mo; Ipsen, Paris, France; n = 1). The patients continue to use long-acting SSAs in between PRRT courses; this treatment was stopped at least 6 weeks before radioligand treatment was administered.

All patients presented with an Eastern Cooperative Oncology Group performance score of 0-1,12 and received PRRT (⁹⁰Y-DOTATOC, ¹⁷⁷Lu-DOTATOC/-DOTATATE, or both) for progressive disease observed within 1 year before start of treatment, despite treatment with SSAs. The decision on progressive disease was made based on increase in tumor burden-related complains, progressive increase in tumor markers (gastrin and chromogranin A) levels, as well as on objective increase in tumor mass on imaging modalities (as observed by a single appointed radiologist at each center).

Clinical Assessment

To define symptomatic tumors, patient files were reviewed to look for the presence of symptoms of gastrinomas, such as diarrhea, weight loss, abdominal pain, and/or symptoms associated with severe peptic ulcers or gastro-esophageal reflux (eg, heartburn, epigastric/retrosternal pain, nausea, bloating, or vomiting). The presence of other disorders (eg, other endocrine diseases/tumors or paraneoplastic syndromes) in patients or family members was also recorded.

Biochemical Evaluation

All patients underwent a complete biochemical assessment, including complete blood count and kidney function tests (as inclusion criteria for considering PRRT).

Serum gastrin and chromogranin A (CgA) were measured in the morning, after an overnight fast, and thereafter at regular intervals (3-6 months) during the study period. Due to the increased risk of severe peptic disease as a result of hypergastrinemia-induced gastric acid hypersecretion, all patients were treated with PPIs during the study follow-up. PPI treatment was not discontinued before blood samples were taken. Serum gastrin levels were measured using the radioimmunoassay kit DiaSorin (Stillwater, MN; normal, 40-108 mU/L); serum CgA levels were assessed using CGA-RIACT (CISBIO International, France; normal, 19.4-98.1 ng/mL).

Functional and Radiological Assessment

Before PRRT, all patients underwent ¹¹¹Indium (¹¹¹In)-pentetreotide scintigraphy (octreoscan) or Ga⁶⁸-DOTATOC/DOTANOC/positron emission tomography (PET)-computed tomography (CT) to determine the presence of somatostatin receptors on tumor cells and patient suitability for PRRT. All patients underwent endoscopic ultrasound, triphasic CT of the abdomen, or magnetic resonance imaging (when CT was contraindicated) to assess disease extension. The imaging data were reviewed by a single appointed radiologist.

Histopathological Assessment

Data from histopathological reports was revised to evaluate the size and the extension of the lesions, signs of invasion (perineural, intravascular or adipose tissue invasion), regional lymph node involvement, and presence of necrosis. Tumor sections immunostained for neuroendocrine markers (chromogranin, neuron-specific enolase, synaptophysin), and for the Ki-67 index of proliferation have been recorded. The diagnosis of gastrinomas was confirmed via tumor morphological analysis together with positive immunohistochemical staining for neuroendocrine markers in the context of appropriate clinical picture and increased gastrin levels in the blood.

PRRT Protocols

[⁹⁰Y-DOTA]-D-Phe1-Tyr3-octreotide (⁹⁰Y-DOTATOC)

The ⁹⁰Ylabeled, 1,4,7,10-tetra-azacyclododecane N,N′,N′′,N′′′-tetraacetic acid (DOTA)-modified SSA Tyr³-octreotide (TOC) is generally synthesized in a 5-step synthetic procedure according to good laboratory practice.13 For radiolabeling, kits containing 220 μg DOTATOC, 18.3 mg ascorbic acid, and 160 mg sodium ascorbate are labeled with 3.7 GBq/m² body surface ⁹⁰Y (β-emission for therapeutic purpose) and 111 MBq ¹¹¹In-Cl (γ-emission for intratherapeutic imaging). The solution is heated at 95°C for 25 minutes, and quality control is usually performed using solid phase extraction and/or high performance liquid chromatography, with a minimum required labeling purity >99.5%.

¹⁷⁷Lu-[DOTA⁰,Tyr³,Thr⁸]-octreotide (¹⁷⁷Lu-DOTATATE)

[DOTA⁰,Tyr³]octreotate kits are usually prepared with 120 μg [DOTA⁰,Tyr³]octreotate, 37.8 mg sodium ascorbate, and 7.5 mg gentisic acid in 300 μL 0.05 M HCl.¹⁴ Kits are stored at −20°C until use. For radiolabeling, ¹⁷⁷LuCl₃ is diluted in 0.05 M HCl at a concentration of 11.1 GBq/mL, and 2220 MBq ¹⁷⁷LuCl₃ is added to each kit. The mixture is heated for 30 minutes at 80°C. The labeling yield is subsequently checked using instant thin-layer chromatography with 0.1 M Na citrate (pH 5.0) as a solvent. The labeled peptide migrates from the origin till Rf = 0.67, whereas the free radionuclide migrates with the solvent front (Rf = 1). The radiochemical purity is determined via high-performance liquid chromatography. The required labeling yield is >98%.

The intratherapeutic biodistribution of the radiopeptide was assessed using planar whole-body imaging in all 1556 patients treated in Basel as described.15, 16 The maximal tracer uptake in the tumor was visually scored by a panel of 3 board-certified nuclear medicine physicians blinded to each patient's baseline and follow-up results using a 4-point scale17 (0 = no uptake, 1 = uptake lower than liver uptake, 2 = uptake similar to liver uptake, 3 = uptake higher than liver uptake.

PRRT (either ⁹⁰Y-DOTATOC or ¹⁷⁷Lutetium-DOTATOC/DOTATATE) were administered as follows: peritherapeutic nausea prophylaxis was given using Zofran (ondansetron) and Fortecortin (dexamethasone). An infusion of 1000 mL physiological NaCl solution containing 20.7 mg/mL arginine and 20.0 mg/mL lysine was started 30 minutes before and continued for 3 hours after radiopharmaceutical injection to inhibit the tubular reabsorption of the radiopeptide.

Evaluation of Response to PRRT

Tumor response to PRRT was defined according to the Response Evaluation Criteria in Solid Tumors.18 Patients were considered in complete response if symptoms disappeared, serum markers returned to normal, and all target lesions disappeared on imaging studies. Patients were considered in partial response if at least 30% reduction in the sum of the longest diameter of target lesions was observed. Patients were considered in progressive disease if there was a ≤20% increase in the sum of diameters of target lesions or if new lesions were noticed; and stable disease if neither sufficient shrinkage nor sufficient increase were observed.

Statistical Analysis

Results are expressed as the mean ± SD. Differences in variables were analyzed using a chi-square test and Student t test using SigmaStat 2.03 (Systat Software Inc., Point Richmond, CA). P < .05 was considered statistically significant.

RESULTS

The clinical characteristics of all patients included in the study are shown in Table 1. From a total of 154 consecutive patients with neuroendocrine tumors from both centers which were treated with PRRT for progressive disease during the study period, 11 patients with gastrinoma (7%) received the radiolabeled treatment. The cohort included 7 men and 4 women with a median age of 58.0 years. The mean duration of follow-up was 42 months (range, 18-72 months). Acid-related peptic complaints (eg, heartburn, epigastric pain, nausea, bloating, vomiting) were present in all patients, whereas hypergastrinemia- or tumor burden-related symptoms (eg, diarrhea, weight loss, vomiting, weakness, limb edema, ascites, abdominal fullness, jaundice) were observed in 8 patients (73%). In 2 patients (18%), pluri-hormonal secretion was diagnosed (in 1 patient, the tumor secreted both gastrin and ACTH, inducing a para-neoplastic Cushing syndrome; in the other patient, the tumor secreted both gastrin and glucagon, but with a predominant clinical picture of gastrinoma, in terms of severe and long-lasting peptic disease, and diarrhea). The disease was metastatic in all patients, with the most frequent sites of spread being the liver (37%), liver and bones (18%), liver and spleen (9%), liver and abdominal cavity (27%), and liver and lymph nodes (9%). Six of 11 patients (55%) had previously undergone surgical exploration of abdominal cavity, and palliative excision of the primary tumor (Whipple operation) was performed in only 2 patients (18%); 1 patient underwent transarterial chemoembolization (9%). In all patients, the diagnosis was of sporadic gastrinomas.

Table 1. Demographic and Clinical Characteristics of the Study Patients

Characteristics	All Patients (N=11)
Age, y, median	58.0
Men/Women	7/4
Caucasians	11 (100%)
Diameter of primary tumor, cm, mean±SD	8.10±4.9
Localization, pancreas/pancreas and duodenum	10/1
Symptoms, diarrhea, weight loss, weakness, limb edema	8 (73%)
Peptic disease	11 (100%)
Multiple primary tumors	1 (9%)
Metastatic disease	11 (100%)
Surgical excision of primary tumors	6 (55%)

Basal Evaluation (Before Treatment)

CT defined tumor burden in all 11 patients, showing a diameter of 8.1 ± 4.9 cm for the largest lesion (range, 3-19.6 cm). The tumors were single and localized in the pancreas in 10 (91%) patients and multiple (≥2 tumors on CT) (pancreatic and duodenal) in 1 (9%) patient.

Functional imaging using octreoscan or Ga⁶⁸-DOTATOC/DOTANOC/PET-CT demonstrated a high to intense tumor uptake that was at least as high as the uptake in the normal liver tissue (score 2-3) in all patients.17

Histopathological assessment (using previous surgical excision samples or endoscopic ultrasound-guided trucat biopsies) demonstrated well-differentiated neuroendocrine grade 1 carcinomas (Ki-67 index ≤2) in 8 patients and grade 2 carcinomas (Ki-67 index 3-20) in the other 3 patients.19 The mean Ki-67% proliferation index was 4.4% (range, 1%-15%).

Gastrin and CgA levels were elevated in all patients, reaching levels of 4831 ± 825 mU/L for gastrin (normal, 40-108 mU/L) and 2400 ± 352 ng/mL for CgA (normal, 19.4-98.1 ng/mL).

Follow-up Assessment (After PRRT)

The type of PRRT used was based on the number, size, and distribution of the tumor lesions. Usually, ⁹⁰Y was employed for patients with bulky disease, whereas ¹⁷⁷Lu was administered in patients with miliary disease, as well as with renal compromise. Two of 11 patients received ¹⁷⁷Lu-DOTATATE, whereas 9 of 11 patients received alternately ⁹⁰Y and ¹⁷⁷Lu-DOTATOC (Table 2). PRRT was administered in divided intravenous doses; treatment intervals were 6-10 weeks. The mean normalized cumulative dose reached 411.5 ± 235.4 mCi/m² for ⁹⁰Y-DOTATOC. For ¹⁷⁷Lu-DOTATOC/DOTATATE, 130-200 mCi doses were used, with a final cumulative dose of 760 mCi (Table 2). The mean number of PRRTs was 2 ± 1.26 courses per patient (range, 1-4) based on tumor uptake assessment performed before and after the first treatment.

Table 2. Characteristics of Tumor Response to PRRT in Study Patients and Details Regarding Cumulative Doses of Radionuclide Treatment

Patient Characteristics				^90Y-DOTATOC		^177Lu-DOTATATE^*/DOTATOC		Response to PRRT
Patient No.	Age, y	Sex	SSA Dosage (mg/mo)	Cumulative Doses, mCi	No. of Courses	Cumulative Doses, mCi	No. of Courses	CR	PR	SD	PD (After Initial Response)
^90Y indicates ^90Yttrium; ^177Lu, ^177Lutetium; SSA, somatostatin analogue; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease. a Two patients receiving ^177Lu-DOTATATE.
1	50	F	Sandostatin LAR (30)	470	2	135a	1	−	+	−	+ (died)
2	51	M	None	—	—	760	4	−	−	+	−
3	59	F	Sandostatin LAR (30)	330	2	—	—	−	+	−	−
4	58	M	Sandostatin LAR (30)	170	1	—	—	−	+	−	−
5	42	F	None	900	4	—	—	−	−	+	+ (died)
6	58	M	Sandostatin LAR (20)	550	3	—	—	−	−	+	+ (died)
7	58	M	Sandostatin LAR (30)	600	3	130a	1	−	−	+	+ (died)
8	81	M	Sandostatin LAR (30)	400	2	184	1	−	−	+	−
9	59	M	Somatuline Autogel (120)	170	1	135	1	+	−	−	−
10	84	F	Sandostatin LAR (30)	135	1	—	—	−	+	−	−
11	39	M	Sandostatin LAR (30)	390	2	—	—	−	+	−	−

Symptomatic improvement appeared in all patients, in terms of disappearance of diarrhea and abdominal pain and significant amelioration of heartburn, nausea, and vomiting; all patients reported amelioration in the tumor burden-related symptoms. Patients' weight increased from 72.2 ± 15.8 kg to 74.8 ± 17.1 kg (P < .05).

Serum gastrin decreased in all patients from 4831 ± 825 mI/L before PRRT to 933 ± 104 mI/L at last follow-up (normal, 40-108 mI/L; P < .005) (Figure 1A). The levels of serum CgA were consistently elevated before PRRT up to 2400 ± 352 ng/mL and decreased significantly to 860 ± 69.8 ng/mL (normal, 19.4-98.1 ng/mL; P < .005) (Figure 1B). In parallel with the clinical benefit, the treatment significantly improved the levels of liver alkaline phosphatase and of γ-glutamyl transferase (Table 3).

Figure 1. Serum levels of (A) gastrin (normal, 40-108 mI/L) and (B) CgA (normal, 19.4-98.1 ng/mL) before and at last follow-up are shown for all patients included in the study (P < .005).

Table 3. Biochemical Changes Observed Before and After PRRT in the Study Patients

Variable	Normal Range	Before PRRT	Immediately After PRRT	At Last Follow-up	Pa
		Mean±SD	Mean±SD	Mean±SD
a The P value for each variable was calculated in relation to the value before PRRT administration.
White blood cell count, K/μL	4800-10,800	6568±1074	3175±1238	5237±2702	<.05
Hemoglobin level, g%	12-16 (women), 13.5-16 (men)	12.37±2.29	12±1.92	11.5±1.56
Platelet count, K/μL	130,000-400,000	246,142±9579	128,285±6182	145,500±3531	<.05
Alkaline phosphatase level, U/L	30-120	462.4±32.9	311.6±37.4	289±42.8	<.05
γ-Glutamyl transferase level, U/L	0-38	386.4±30.8	304.14±28.2	246.6±52	<.05
Creatinine level, mg%	0.5-0.9	0.88±0.46	0.91±0.33	0.94±0.14

The mean diameter of the largest lesion decreased from 8.1 ± 4.9 cm to 5.3 ± 2.1 cm (P < .05) (Figure 2). Periodic radiological surveillance showed complete response at 1 year after the last PRRT in 1 (9%) patient (Figures 3A and 3B), partial response in 5 (45%) patients (Figures 3C and 3D), and tumor stabilization in 5 (45%) patients.

Figure 2. The mean diameter of the largest tumor before and after PRRT is shown as assessed by CT.

Figure 3. CT scans are shown for 2 patients with metastatic gastrinoma before treatment (left) and 6-12 months after the last treatment (right). The lesions are indicated by black arrows. Note the liver lesions as well as primary tumor disappearance (complete response) in patient 1 (A and B) and the decrease in the number and size of liver lesions (partial response) in patient 2 (C and D).

Details regarding the type and dosage of PRRT for each patient are summarized in Table 2. Four of 11 patients received ⁹⁰Y-DOTATOC and ¹⁷⁷Lu-DOTATOC/DOTATATE courses alternately, 6 patients received ⁹⁰Y-DOTATOC alone, and 1 patient received ¹⁷⁷Lu-DOTATOC alone. There was no clear relation between the type or number of PRRTs and tumor response, considering the small number of patients included in the study. However, 10 patients received ⁹⁰Y-DOTATOC, which is considered more appropriate for bulky disease. During follow-up, the antitumor effect of PRRT persisted in 7 (64%) patients for a median period of 14 months (ongoing); 4 (36%) patients died due to tumor progression, with a median TTP of 11 months; in this group, the mean survival time after the last PRRT reached 14 ± 6.9 months. It is noteworthy that patients who eventually died following PRRT because of tumor progression presented with massive metastatic disease involving the liver and abdominal cavity; in these patients, the mean Ki-67 index of tumor proliferation measured 3.75 ± 1.89, significantly higher compared with a Ki-67 of 1.28 ± 0.48 (P < .05) in the subgroup of patients who survived.

Tumor response positively correlated with a higher level of uptake on octreoscan, a limited hepatic tumor mass, and a low Eastern Cooperative Oncology Group performance score. Following PRRT, tumor uptake was higher than liver uptake in all patients (score = 3).

All patients tolerated the treatment well, and no severe adverse effects were observed during the follow-up period. The most frequent adverse effects involved a transient decrease in complete blood count (hematologic toxicity grade 1-2), which was noted in 8 (73%) patients (Table 3), but without significant clinical consequences or need for any therapeutic intervention. No myelodysplastic syndrome was observed. One (9%) patient had a temporary increase in creatinine levels.

DISCUSSION

In gastrinoma patients with advanced inoperable disease, it is important but difficult to decide on appropriate treatment, because alternative therapeutic options (eg, cytoreductive surgery, hepatic artery chemoembolization, chemotherapy) did not prolong life in these patients.20-22 However, in a study including 15 patients with malignant gastrinomas and progressive hepatic involvement, treatment with the SSA octreotide induced an important and long-lasting antigrowth effect in up to 50% of these patients,23 suggesting that octreotide should routinely replace chemotherapy as a standard treatment for patients with progressive gastrinoma.

The increased avidity on functional imaging modalities (octreoscan or Ga⁶⁸-DOTATOC/DOTANOC/PET-CT) demonstrated in most patients with metastatic gastrinomas24 suggests that PRRT may be a logical treatment option in this subclass of NET patients. A recent study involving 129 NET patients treated with ¹⁷⁷Lu-octreotate (including 8 gastrinoma patients)25 demonstrated a favorable effect on tumor size in 47% of all study patients. In this report, all gastrinoma patients seemed to benefit from PRRT: partial remission or minimal response appeared in 7/8 (88%) patients, and tumor stabilization in 1/8 (12%) patients, with no complete response. Consistently, in our study, tumor response was initially observed in all patients, with partial response or stable disease in 10 patients, and complete tumor disappearance in 1 patient. The use of ⁹⁰Y-DOTATOC rather than ¹⁷⁷Lu-DOTATOC/DOTATATE in most of our patients could be more appropriate for bulky disease26, 27 and may explain the somewhat better tumor responses observed in our patients. Although the number of gastrinoma patients included in the present study as well as in the previous one is relatively low, these observations strongly suggest that PRRT may have an important role in the treatment of patients with metastatic gastrinoma, who show a good performance status before therapy initiation.

The reported antitumor effects of ⁹⁰Y-DOTA-octreotide in NET patients differ among various studies: remission rates of 7%-33% in gastro-entero-pancreatic (GEP) tumor patients have been reported, probably as result of differences in the protocols used.14, 28, 29 Furthermore, treatment with ¹⁷⁷Lu-DOTA-octreotate, which has a higher affinity for the subtype 2 somatostatin receptor, induced complete or partial remissions in 30% of patients.30 Tumor regression was positively correlated with a higher level of uptake on octreoscan, a limited hepatic tumor mass, and a high Karnofsky performance score,31 as demonstrated also in our patients. The median TTP after PRRT for different GEP NETs has been reported not to approach 25-32 months32, 33; usually, in these studies progressive disease developed after an initial benefit from regular PRRT. In our gastrinoma patients, the median TTP reached up to 11 months in a subgroup of 4 patients who eventually died due to an aggressive tumor behavior, whereas in the other 7 patients, the antitumor effect of PRRT still persists after a median period of 14 months. Interestingly, a recent study using “salvage” additional cycles with ¹⁷⁷Lu-octreotate for patients with progressive GEP NETs after initial PRRT demonstrated supplementary antitumor effects, with a median TTP of 17 months.34 These observations are indeed encouraging and may suggest that the combination of “cold” SSAs (eg, monthly injections of usual SSAs), as previously suggested,23 and “hot” SSAs (PRRT using radiolabelled SSAs), as in our group, may be more effective than each modality alone in gastrinoma patients with progressive disease.

In the present study, there were no serious adverse effects; renal insufficiency occurred in 1 patient and was transient; however, this may be attributable to the lack of serious predisposing factors for post-PRRT renal failure (namely preexisting renal compromise) as well as the renoprotection via pre-PRRT amino acid infusion. Although myelodysplastic syndrome and leukemia have been infrequently reported as adverse effects after treatment with ¹¹¹In- and ⁹⁰Y-labeled SSAs in relation with high cumulative doses,33 in the present study none of the patients developed these severe complications, probably due to the limited maximum cumulative dose used. However, mild and transient bone marrow suppression was found in two-thirds of our patients, as reported in most studies with ¹¹¹In- and ⁹⁰Y-labeled SSAs.33, 35-37

Based on our study and on other reports,38 treatment with radiolabeled SSAs seems to be a promising tool for the management of patients with progressive metastatic gastrinomas. Symptomatic improvement may occur in most patients, regardless of the type of PRRT (eg, ⁹⁰Y-, or ¹⁷⁷Lu-labeled SSAs), and the results are also encouraging in terms of tumor regression. Because of the limited efficacy of alternative therapies, the significant albeit temporary success rate of therapy with PRRT, and in the absence of serious adverse effects, we recommend its early use in patients with metastatic gastrinomas, before tumor progression and patient deterioration have occurred. Moreover, because “cold” SSAs have been shown to have beneficial antitumor effects in these patients, we believe that the use of combined “cold” and radiolabeled “hot” SSAs may have synergistic antiproliferative activity in patients with metastatic gastrinomas.

CONFLICT OF INTEREST DISCLOSURES

D.J. Gross received support from Novartis and Ipsen and honoraria from Novartis, Ipsen, and Pfizer. S. Grozinsky-Glasberg received support from Novartis.