High risk of myelodysplastic syndrome and acute myeloid leukemia after ¹⁷⁷Lu-octreotate PRRT in NET patients heavily pretreated with alkylating chemotherapy

Dear Editor,

Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) are rare but seen with increasing incidence. Current medical options for the management of nonresectable GEP-NETs include somatostatin analogs, targeted therapies, chemotherapies, and radiological and radionuclide therapies.

Peptide receptor radionuclide therapy (PRRT) is a modern therapeutic approach using radionuclide combined with somatostatin analog peptide whose affinity with somatostatin receptors (SSRs) allows targeting disseminated tumor disease. According to the ENETS guidelines, PRRT is indicated for patients with nonresectable, progressive, grade 1 or 2 GEP-NETs with high uptake on SSR scintigraphy (Pavel et al. 2012). In a large retrospective study of 310 GEP-NETs, 46% had tumor response with ¹⁷⁷Lu-octreotate therapy (Kwekkeboom et al. 2008) with good progression-free and overall survival. In addition, first results of the prospective randomized NETTER-1 trial comparing ¹⁷⁷Lu-octreotate PRRT and octreotate LAR 60 mg have recently showed a median progression-free survival not reached at 25 and 8.4 months, respectively (Strosberg et al. 2015). PRRT is generally well tolerated, short-term side effects include mild fatigue, hematological and renal toxicity. Regarding longer term hematological side effects, myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML) was reported in 0.2–5.4% of the patients in large series (Kwekkeboom et al. 2008, Imhof et al. 2011, Sabet et al. 2013, Kesavan et al. 2014, Bodei et al. 2015). Here, we report a much higher occurrence of MDS or AML in a single-center experience in patients treated with ¹⁷⁷Lutetium-octreotate at late disease stage, after alkylating-based chemotherapy.

Our study included all 20 consecutive patients treated with ¹⁷⁷Lutetium-octreotate PRRT between January 2004 and January 2011 at our center. All patients had progressive metastatic GEP-NETs. PRRT was performed in the Department of Nuclear Medicine of the Erasmus Medical Center of Rotterdam (Rotterdam, the Netherlands) due to the unavailability of PRRT in France. This unavailability explained why relatively few patients received this treatment in our center, and generally after first-line chemotherapy. As defined in previous reports from Rotterdam’s team (Kwekkeboom et al. 2001), the radiochemical purity of ¹⁷⁷Lu-octreotate was 88%, and then reached yield after addition of DTPA (diethylenetriaminepentaacetic acid) approached 99.99%. Patient medical records were reviewed to collect relevant data on demographics, tumor characteristics, surgery, treatments, and tolerability. A Student’s t-test was used to evaluate prognostic variables for the occurrence of therapy-related MDS or AML (t-MDS/AML).

Baseline clinical characteristics of the 20 patients are described in Table 1. Median follow-up from PRRT was 3.1 years (range 0.3–8.9). Treatment with ¹⁷⁷Lu-PRRT consisted of four cycles of 7.5 GBq; 16 patients received the four full dose cycles (one of them received two additional cycles) and 4 received lower dose due to early hematological toxicity (one of them one cycle).

PPRT induced short-term thrombocytopenia, neutropenia, and anemia in 25, 15, and 10%, respectively. However, four (20%) patients developed delayed hematological toxicity with MDS or AML, diagnosed 30, 31, 54, and 70 months, respectively, after PRRT. All patients, before PRRT, had received prior chemotherapy containing an alkylating agent for 20, 12, 6, and 12 cycles, respectively. Three of the four patients (75%) had developed early hematological toxicity (grade 3–4 thrombocytopenia) during PRRT. At the time of MDS/AML diagnosis, the underlying tumor was active in three patients, and three had bone metastases. Three patients had MDS according to the WHO criteria (Della Porta et al. 2015), including refractory cytopenia with multilineage dysplasia in two cases, and refractory anemia with excess blasts type 2 in one case; the last patient had the WHO-defined AML (with, however, only 21% marrow blasts, i.e., partial blast infiltration very close to MDS diagnostic criteria). Karyotype showed monosomy 7 in three cases, with or without other chromosomal deletions, especially monosomy 5, and was not performed in the last patient.

Compared to other patients, the four patients who developed MDS or AML had received, prior to PPRT, more cycles of chemotherapy (mean: 13.8 vs 4.7, P=0.001) and more cycles of alkylating agents (mean: 12.5 vs 3.75, P=0.001), and had more frequently experienced early hematological toxicity (75 vs 13%, P=0.03) (Table 2). Two of the patients received supportive care only and two received azacitidine. Survival was 6, 2, 2, and 54 months, respectively. During the study period (2004–2011), 95 additional patients with GEP-NETs were treated with alkylating-based chemotherapy without subsequent PPRT at our center. Only one (1%) developed MDS.

The high rate of MDS or AML (20%) we report in this limited series of 20 nonresectable NETs treated with ¹⁷⁷Lu-PPRT after heavy pretreatment with chemotherapy is therefore much higher than in large published series of PRRT (Kwekkeboom et al. 2008, Imhof et al. 2011, Sabet et al. 2013, Kesavan et al. 2014, Bodei et al. 2015). This higher rate may be due to the fact that most of our patients had received chemotherapy and alkylating agents prior to PPRT, compared to previously published series of PPRT in metastatic NETs, where less than one-third of the patients had received chemotherapy before PRRT. Indeed, in our series, patients who developed MDS or AML had received more chemotherapy and more alkylating agents than other patients (P=0.001). Reasons why patients had received more chemotherapy prior to PRRT in our series probably include the fact that PPRT was until recently unavailable in France, requiring a partnership with centers abroad. Thus, PRRT was not a frontline therapy for metastatic GEP-NETs in France, where patients first received conventional systemic chemotherapy, targeted therapies, and transarterial chemoembolization.

In another study, about 65 patients receiving ¹⁷⁷Lu-octreoate combined with capecitabine (n=28) or capecitabine and temozolomide (n=37), no MDS/AML was reported in the former group, but two (5.4%) cases were reported in the latter group (Kesavan et al. 2014). Temozolomide is an alkylating agent, further supporting our assumption that the risk of MDS/AML, very low with ¹⁷⁷Lu-octreotate PRRT alone and possibly ¹⁷⁷Lu-octreotate PRRT combined with nonalkylating chemotherapy, may be higher when ¹⁷⁷Lu-octreotate PRRT is combined with alkylating agents.

The duration of follow-up in our series (3.1 years) cannot explain differences of MDS/AML rates compared with other studies: indeed, although three studies had shorter follow-up (1.6, 1.9, and 2.6 years) (Kwekkeboom et al. 2008, Imhof et al. 2011, Sabet et al. 2013), two other studies had similar or longer follow-up (3.0 and 4.2 years) (Kesavan et al. 2014, Bodei et al. 2015). Because of the known leukemogenic role of alkylating agents, we cannot exclude that MDS/AML in our series was a direct consequence of prior chemotherapy. Indeed, latencies before the first cycle of chemotherapy, and the diagnosis of the hematological malignancies were 12.6, 9.8, 3.8, and 10.4 years, respectively. It is well known that the maximum risk of MDS/AML ranges from 3 to 10 years after the myelotoxic treatment. However, the risk of MDS/AML in GEP-NETs treated with alkylating-based chemotherapy alone may be relatively small: during the study period (2004–2011), 95 additional GEP-NETs patients were treated with this approach without subsequent PPRT at our center, and only one (1%) developed MDS.

Other risk factors for MDS/AML, in our study, included early hematological toxicity after PPRT, observed in 3 of the 4 patients who subsequently developed MDS/AML, compared to only 2 of the 16 patients who did not develop MDS/AML. Platelet toxicity was also significantly related to development of secondary MDS or AML in a previous study (Bodei et al. 2015): close hematological monitoring therefore seems recommended in patients experiencing early hematological toxicity after PRRT. MDS and AML cases observed in this series were typical of those observed after alkylating agents or purine analogs. Indeed, they all had partial marrow blast infiltration (maximum 21%) and the three karyotyped patients had monosomy 7.

The impact of bone metastases in the occurrence of MDS/AML remained unclear, and it had no significant impact in our study (P=0.39). In a pilot study in 11patients with GEP-NETs and florid bone metastases (with advanced widespread metastatic bone disease) receiving PRRT (Sabet et al. 2014), a higher than usual rate of hematological toxicity was reported, with grade 3–4 reversible hematotoxicity of in 35% of the patients, but delayed hematological toxicity was not reported. In other studies, no statistical correlation between bone metastases after PRRT and the recovery of acute hematotoxicity or the subsequent occurrence of MDS/AML was seen (Sabet et al. 2013, Bergsma et al. 2015).

In conclusion, our study, although based on a limited number of patients, suggests a high occurrence of MDS/AML in patients with GEP-NETs treated with PRRT after previous chemotherapy with alkylating agents. The indication of PRRT should take into consideration the importance of previous chemotherapy. Exposure to alkylating agents should be avoided in patients with low-grade NET, who have a long survival expectancy and a significant likelihood of benefiting from PRRT, because it may compromise the safety and future applicability of this more effective therapy. Regular and prolonged monitoring of blood counts is mandatory, especially in patients experiencing early hematological toxicity after PRRT.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This research did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.