Introduction

Pancreatic cancer remains a devastating health problem, with an estimated 28,900 deaths in 2001 and a 5-year survival rate of 4% [1]. The disease is characterized by early locoregional spread and distant metastases. As a consequence, the majority of patients (pts) present with advanced disease which is not resectable. For these pts, systemic chemotherapy has been largely ineffective [2].

In this palliative setting, gemcitabine (Gem) has been the standard treatment for advanced pancreatic cancer. More than a decade ago, it had already been shown to result in a superior clinical benefit, response and survival compared with bolus 5-fluorouracil [3]. Since then, clinical trials have explored the combination of Gem with other cytotoxic or the biologically “targeted” agents. However, promising results in small phase II trials have not translated into survival improvements in larger randomized phase III trials in the advanced disease setting.

Recently, modest survival improvements with the use of combination treatment with Gem and erlotinib have been reported [4]. Still, the median survival of pts with advanced disease continues to be less than 6 months, and evaluation of novel therapeutic targets is needed to improve the outcome for these pts.

The epidermal growth factor receptor (EGFR) seems to play a particularly important role in the carcinogenesis of human cancers [5], including pancreatic cancer. Increased expression of EGFR and its ligand has been detected in human pancreatic cancer tissue [6]. Moreover, coexpression of EGFR and its ligand has been proven to predict poor prognosis in pancreatic cancer [6], and it has been proposed that coexpression of EGFR and its ligand functions as an autocrine loop to constantly stimulate cell proliferation. Therefore, different strategies have been developed to inhibit the aberrant EGFR-associated signal transduction cascade. The most important therapeutic approaches include small-molecule tyrosine kinase inhibitors [7], which act by interfering with ATP binding to the receptor, and monoclonal antibodies [8], which bind specifically to the extracellular region of the receptor, inhibiting its dimerization and autophosphorylation.

Nimotuzumab, also known as h-R3, is a humanized anti-EGFR monoclonal antibody developed at the Center of Molecular Immunology in Havana, Cuba. Nimotuzumab inhibits EGFR by binding to domain III of the receptor’s extracellular region, partially blocking the EGF binding site, as well as stabilizing the receptor’s protein conformation unfavorable to dimer formation [9]. As a consequence of diminished EGFR phosphorylation and decrease in growth and cell division commands, nimotuzumab reduced proliferation in vitro and in vivo. It has been suggested that nimotuzumab’s in vivo cytotoxic and pro-apoptotic effects are mainly mediated by a decrease in VEGF production [10]. Furthermore, nimotuzumab promotes an in vivo antineoplastic effect not dependent on inactivating EGFR because it can stimulate the cell- and complement-mediated cytotoxic immune responses to attack the tumor cells to which it is bound [11, 12]. In preclinical studies, the antibody has shown potent antitumor activity [10, 12, 13].

Based on phase I data, the recommended dose has been established at 200 mg weekly [14]. Nimotuzumab has been approved in several countries for the treatment of head and neck tumors [15, 16] and glioma [17]. A previous phase II study in children with high grade brain tumors showed activity of nimotuzumab as a monotherapeutic agent, even in the prognostically very unfavorable pontine gliomas. No clinically relevant drug-related side effects were reported [18].

We report the results of a multicenter phase II trial of nimotuzumab monotherapy for pts with advanced pancreatic cancer. The objectives of the trial were to determine the objective tumor response rate, progression-free survival (PFS), overall survival (OS) as well as the safety profile.

Patients and methods

Study design and treatment

This was a multicenter single-arm phase II study evaluating the efficacy and safety of nimotuzumab monotherapy in pts who failed standard chemotherapy with Gem or another first line regimen for advanced disease. Pts having at least one measurable lesion as defined by RECIST criteria were eligible for the study. Nimotuzumab was given intravenously as induction therapy at 200 mg once weekly for 6 weeks (wks). Follow up by computed tomography (CT) scan was performed after 8 weeks. Pts continued receiving treatment 3-weekly until death, disease progression (PD), occurrence of intolerable side effects, investigator’s decision, or withdrawal of consent, whichever came first. Following discontinuation of treatment for reasons other than withdrawal of consent, death or PD, pts were followed every 6 weeks.

Endpoints included tumor response (RECIST), PFS and safety. Blood samples were collected prior to the first dose, at the end of the infusion, and 3 h, 6 h, 48 h as well as every time before subsequent nimotuzumab doses were administered. Nimotuzumab concentrations in the serum were measured by cellular ELISA and calculations performed using the Kinetica program.

Patient eligibility

Pts were eligible if they had histologically or cytologically confirmed metastatic adenocarcinoma of the pancreas. Pts had to be aged over 18 years. Further eligibility criteria were: Karnofsky Performance Status over 70%; adequate organ function; and an expected survival of at least 3 months. Written informed consent was obtained from all pts. Pts had to have measurable disease according to RECIST criteria (i.e., lesions that can be accurately measured in at least one dimension with the longest diameter 20 mm using conventional techniques or 10 mm using spiral CT scan).

Toxicity and efficacy assessment

Pts were monitored for safety using adverse event information, physical examinations, the Karnofsky Performance Status, vital signs, and laboratory evaluations. Adverse events were graded using National Cancer Institute Common Toxicity Criteria version 3.0 (CTC v3.0).

On-site tumor assessments were performed by the investigator and local radiologist, and the results of those assessments were utilized to determine whether to continue or discontinue treatment. Following discontinuation of study treatment for reasons other than withdrawal of consent or PD, pts were followed for tumor response every 6 weeks until disease progression or initiation of new anticancer therapy.

Statistical considerations

Primary endpoint of the study was the remission rate after administration of at least 6 doses of study drug (RR6). The study was designed as a non-randomized non-blinded multicenter phase II trial to distinguish between the null hypothesis H0: RR6 ≤ p0 = 0.10 versus the alternative H1 : RR6 > p1 = 0.10.

Secondary endpoints were PFS, safety and control of leading symptoms including pain. The primary endpoint RR6 is defined as the ratio \( {\hbox{RR6}} = {{\hbox{n}}_{\rm{RR6}}}/{{\hbox{n}}_{\rm{FA}}} \) where nFA = number of pts eligible for response according to RECIST criteria after administration of at least 6 doses of study drug (full analysis set) and nRR6 = number of pts of the full analysis set with complete or partial response. The primary efficacy variable RR6 was planned to be analyzed by a one-sided exact binomial test for proportions of testing H0 at the level of alpha = 0.05.

Results

Patient demographics

A total of 54 pts (28 women/26 men; ECOG status of 1 [n = 40] or 0 [n = 14], median age 63.6 years [range 46–83 years]) were enrolled from 5 sites in Germany. A total of 7 pts had locally advanced disease only, the majority (47 pts) presented with metastatic disease. All pts received prior cytotoxic treatment, 1 regimen (28 pts), 2 regimens (20 pts) or 3 or more different regimens (6 pts) (Table 1).

Table 1 Patient and tumor characteristics (N = 54 pts)
Full size table

Safety

In general, nimotuzumab monotherapy was well tolerated. Treatment-related adverse events with reported possible relation to the study drug were CTC-grade 3 for a total of 3 pts: a vascular event (deep vein thrombosis), one patient with hemorrhage/bleeding and complete recovery, and one patient with gastrointestinal obstruction. Other adverse events with a possible relation to nimotuzumab (CTC-grade 1/CTC-grade 2) were constitutional (fever, chills) in 7 pts (5/2), skin toxicity (rash) in 5 pts (CTC-grade 1 only), fatigue in 5 pts (4/1), nausea in 4 pts (2/2), pain in 2 pts (CTC-grade 1 only), allergic reactions in 2 pts (CTC-grade 2 only), and other adverse events in one patient each (Table 2).

Table 2 Treatment-related adverse events with possible relation to study drug
Full size table

No patient was discontinued from nimotuzumab treatment because of adverse events. A total of 29 pts (53.7%) were discontinued due to objective progression of the disease, and 19 pts (35.2%) due to clinical progression of the disease prior to the first follow-up by CT.

After a single dose of 200 mg, the mean value of Cmax was calculated to 141 ± 33 μg/ml. The t1/2 was calculated to 45 h, volume of distribution to 1.46 ± 0.3 l, respectively. The total clearance was determined to 23 ± 6 ml/h. The trough values after 168 h were 6.2 ± 6.3 μg/ml.

Efficacy results

A total of 35 pts were evaluable for objective response (Table 3). No objective response was seen. However, a total of 6 pts showed sustained stabilization of previously progressive disease with a median duration of disease control of 19.2 weeks (95% CI 14.1–26.1 weeks) in these pts. A correlation between the occurrence of skin toxicity and response could not be established. All 54 pts were evaluable for Kaplan-Meier estimates of the median PFS and OS. The median OS was 18.1 weeks (95% CI 16.3–19.8) (Fig. 1a). The median PFS for all pts was 6.7 weeks (95% CI 6.3–7.0 weeks) (Fig. 1b). The 1-year PFS rate was 10.3% (95% CI 4.5–23.7%). Karnofsky Performance Status and body weight did not change in 43 out of 54 (79.6%) evaluable pts during the first 6 weeks after first administration of nimotuzumab.

Table 3 Antitumor efficacy
Full size table
Fig. 1
figure 1

a Overall survival (OS) in weeks. b Progression-free survival (PFS) in weeks

Full size image

Five pts (9.2%) received a further line of chemotherapy, mostly 5-FU based schedules. However, none of these pts showed any objective response or clinically benefit on subsequent treatment regimens.

Discussion

Most pts with adenocarcinoma of the pancreas present with locally advanced or metastatic disease. Although single agent Gem is widely accepted as first-line therapy, there is no current standard second-line regimen for advanced pancreatic cancer after Gem failure and there is a paucity of trials in this setting [19, 20].

For pts with adequate performance status, the CONKO-003 trial helped to establish the superiority of second line chemotherapy versus best supportive care [21]. The CONKO-003 trial also demonstrated an OS benefit with the addition of oxaliplatin to infusional 5-FU and leucovorin (OFF regimen). The study showed an improved OS by 2 months in the OFF arm (4.8 months versus 2.3 months, P = 0.0077). Although both regimens were tolerable, pts in the OFF arm experienced significant more neuropathy. There was a significant prolongation of PFS in the treatment arm (13 weeks versus 9 weeks).

However, very few pts who experience disease progression on or after first line treatment will receive second line chemotherapy. This study on locally advanced or metastatic pancreatic cancer was initiated to evaluate the activity and safety of nimotuzumab as 2nd—line therapy. However, nearly half of the enrolled pts received at least 2 prior treatment regimens for advanced disease.

Nimotuzumab binds within an area that overlaps with both the surface patch recognized by cetuximab and the binding site for EGF, while still allowing the receptor to adopt its active conformation, hence warranting a basal level of signalling [9, 22]. Nimotuzumab exhibits different pharmacokinetic properties when compared with other anti-EGFR antibodies [15]. Clinical trials with this antibody, involving >4,000 pts worldwide, have shown evidence of efficacy in the treatment of pts bearing advanced epithelial-derived tumors with low toxicity and without provoking skin rash [14, 15, 23, 24]. Several EGFR antagonists have been approved in various clinical settings, including antibodies against the extracellular domain of the receptor, such as cetuximab, panitumumab, or nimotuzumab, and small tyrosine kinase inhibitors such as erlotinib and gefitinib [25–27]. However, skin rash occurs in the majority of pts treated with these agents and causes a decrease in the quality of life as well as treatment interruptions. The only, and challenging, exception was nimotuzumab.

Our study confirms the favorable toxicity profile of nimotuzumab. Only 3 out of 54 treated pts experienced CTC-grade 3 adverse events, which are more likely to be disease—related than causally linked to the applied study-drug. Other adverse events were generally mild and clinically easy to manage. Skin toxicity occurred in a minority of pts and with CTC-grade 1 only.

The lessened side effects of this monoclonal antibody are probably related to its unique binding affinity and density dependence. A lower affinity between nimotuzumab and EGFR allows for an optimal dose of the drug that is below the toxic dose. Mathematical models predict that the binding affinity (Kd) for anti-EGFR monoclonal antibodies should be in the range of 10-8 M–10−9 M to maximize tumor cell targeting while minimizing normal cell toxicity [15]. Although nimotuzumab is within this range, cetuximab and panitumumab have binding affinities more than 10-fold stronger. The more recently proposed receptor density model, in which nimotuzumab is described as being transiently bound monovalently and strongly bound bivalently to EGFR epitopes [28], further explains nimotuzumab’s diminished adverse events. In normal cells (e.g., skin epithelial cells) EGFR expression is too low to cause nimotuzumab binding in a bivalent manner, thus limiting the antibodies’ potency, and avoiding unwanted toxicities. Tumor cells overexpressing EGFR, on the other hand, have enough receptor density for nimotuzumab to bind bivalently and to robustly inhibit the receptor. This theory is supported by comparative dose binding curves of nimotuzumab and cetuximab on normal skin and kidney cells where nimotuzumab displays significantly less binding than cetuximab [29].

Primary endpoint of the study was the remission rate after administration of at least 6 doses of study drug. Considering the fact that all pts were pretreated, nearly half of them even with two or more regimens, objective tumor response with nimotuzumab as monotherapy could not be expected. However, sustained stabilization of previously progressive disease with a median PFS of 19.2 weeks in 6 out of 54 pts indicates antitumor activity. In addition, median PFS, PFS after 1 year and median OS are comparable to other single-drug regimens for advanced pancreatic cancer after Gem failure.

Currently the EGFR status of tumors as a predictive factor for anti-EGFR directed drugs is under discussion. In this study, immunohistochemical analysis of EGFR expression in the tumors was not performed. More recent analyses in lung cancer have suggested that the presence of EGFR mutations or EGFR amplification measured by fluorescence in situ hybridization may be more useful [30]. However, studies of more than 200 pancreatic tumors did not identify any EGFR mutations [4].

To further evaluate and improve the efficacy of nimotuzumab in pancreatic cancer, a randomized, placebo-controlled multi-center trial in combination with Gem was started.