This article is more than 5 years old. Information may no longer be current.
Management of pancreatic neuroendocrine tumors remains ‘complex, challenging’
Click Here to Manage Email Alerts
The management of pancreatic neuroendocrine tumors remains a complex and challenging endeavor.
Although patients with these rare tumors have a substantially improved prognosis compared with those with most other solid tumors, their disease — once metastatic — is ultimately progressive and fatal. The biology of the disease is variable between patients, and the mainstay of initial therapy remains expectant observation in those patients with small-volume, asymptomatic disease with little or no growth.
Once clinically progressing, slowly progressive tumors can be treated with relatively nontoxic somatostatin analogs such as long-acting octreotide or lanreotide (Somatuline Depot, Ipsen) — either, but not both, as one would not be expected to rescue breakthrough from the other. More rapidly progressive disease, when confined to the liver, is often managed with local therapies such as hepatic arterial embolization.
Leonard Saltz
In experienced hands, these procedures can offer excellent palliation and, although requiring hospitalization and entailing some risk, do not require long-term exposure to systemic agents that might have substantial side effects. More recently, data have shown that either sunitinib (Sutent, Pfizer), a VEGF tyrosine kinase inhibitor, or everolimus (Afinitor, Novartis), an mTOR inhibitor, can be used for disease stabilization or delay of progression; however, substantial tumor shrinkage is rare.
In the trial conducted by Hobday and colleagues, the mTOR inhibitor temsirolimus (Torisel, Pfizer) was combined with the anti-VEGF monoclonal antibody bevacizumab (Avastin, Genentech). The results in this non-randomized phase 2 study were striking — the major objective response rate was 41%. In addition, the waterfall plots indicate that the overwhelming majority of patients experienced some degree of tumor shrinkage. Interpretation of PFS and OS in a non-randomized cohort is, of course, very difficult, but tumors don’t shrink spontaneously, and the degree of tumor regression in this trial is an unequivocal indicator of biologic activity.
One cannot ignore, however, that substantial toxicity which was incurred. Grade 3 (serious) hypertension occurred in 21% of patients, and an additional 32% developed hypertension of the lower grade. Perhaps more importantly, 16% of patients developed grade 3 or greater fatigue, and 63% reported lower-grade fatigue. For a regimen taken over a long period of time, this is very difficult. Finally, mucositis — a known serious toxicity of mTOR inhibitors — was present in 59% of patients, 7% of whom had severe mucositis.
As is typical of current publications, the duration of toxicity is not reported in this manuscript. This information would be important to understand in terms of the acceptability of these toxicities to patients. A short period of grade 3 toxicity would likely be far more acceptable to most patients than a prolonged period of grade 2 toxicity. Even prolonged grade 1 toxicity of events such as fatigue or mucositis could have substantially negative consequences on a patient’s quality of life. The extent of the overall duration of even mild to moderate toxicities needs to be carefully considered when selecting a therapy.
Where do we go with the data reported by Hobday and colleagues? One important direction is already being pursued by the investigators. They note that single-agent bevacizumab has not yet been explored in pancreatic neuroendocrine tumors, and as such, we cannot say whether the activity seen in this trial reflects a meaningful interaction between the two agents, or could much — if not all — of the activity be obtainable from bevacizumab alone? The investigators are therefore now extending their study to a cohort using single-agent bevacizumab, and the results of this ongoing investigation will be important.
Temsirolimus is undoubtedly responsible for a considerable amount of the reported toxicity. Although unlikely to be responsible for the hypertension, much of the fatigue and mucositis is likely attributable to the mTOR inhibitor. In addition, both agents are expensive (and, if bevacizumab is effective at 10 mg/kg, does anyone really worry that 5 mg/kg — the dose used in colorectal cancer — would really be ineffective?). Also, at the time of this writing, neither has either registration or a compendium listing for use in pancreatic neuroendocrine tumors. As such, at present neither agent would be reimbursable for Medicare recipients and likely therefore would not be covered by other third-party payers. Additional information and supportive data will be necessary before we know whether these agents can and should be incorporated into standard practice.
It should be noted that older studies had tended to lump pancreatic neuroendocrine tumors with carcinoids, based on similar morphologic appearances and growth patterns. It is now clear that these diseases are quite different biologically, and that data showing that a therapy that is appropriate for one may not necessarily be extrapolatable to the other. Specifically, pancreatic neuroendocrine tumors appear to be far more sensitive to a number of management options. Thus, activity such as that seen by Hobday and colleagues in pancreatic neuroendocrine tumors should not be extrapolated to carcinoid tumors until and unless specific clinical trials in carcinoid tumors justify such an extrapolation.
For more information:
Leonard Saltz, MD, is chief of the Gastrointestinal Oncology Service and head of the colorectal surgery section at Memorial Sloan Kettering Cancer Center. He also is a HemOnc Today Editorial Board member. He can be reached at saltzl@mskcc.org.
Disclosure: Saltz reports an advisory board role with Genentech/Roche.