A randomised phase II trial of the Polo-like kinase inhibitor BI 2536 in chemo-naïve patients with unresectable exocrine adenocarcinoma of the pancreas - a study within the Central European Society Anticancer Drug Research (CESAR) collaborative network

BACKGROUND

BI 2536, a novel Polo-like kinase 1 inhibitor, was assessed in patients with unresectable advanced exocrine adenocarcinoma of the pancreas.

METHODS

The study employed a two-stage design. Randomised first-line patients received BI 2536 200 mg on day 1 (n=43) or 60 mg on days 1-3 (n=43) every 21 days. Recruitment of second-line patients was planned for a second stage dependent on an interim analysis demonstrating ≥ 2 responses in the first 18 evaluable patients following 12 weeks of treatment and/or tumour control ≥ 12 weeks in 5 patients per schedule. Primary end point was objective response rate (ORR).

RESULTS

By independent review, ORR was 2.3% (all partial) and 24.4% had stable disease as confirmed best response. The second stage was not initiated. Median overall and progression-free survivals were 149 (95% confidence interval (CI), 91-307) and 46 days (95% CI, 44-56). Most common drug-related adverse events were neutropenia (37.2%), leukopenia (29.1%), fatigue (29.1%) and nausea (22.1%); most common grade 3/4-related events were neutropenia (36.0%), leukopenia (27.9%) and thrombocytopenia (8.1%).

CONCLUSION

Given the low ORR and poor survival, further development of BI 2536 monotherapy is not warranted in this population.

Phase i study of the Plk1 inhibitor BI 2536 administered intravenously on three consecutive days in advanced solid tumours

BACKGROUND

This open-label phase i study with an accelerated titration design was performed to determine the maximum tolerated dose of BI 2536, a potent, highly selective small-molecule polo-like kinase 1 (Plk1) inhibitor.

METHODS

Patients with advanced solid tumours received a single 60-minute intravenous infusion of BI 2536 (50-70 mg) on days 1-3 of each 21-day treatment course. Recipients without disease progression or untenable toxicity could receive additional treatment courses. The maximum tolerated dose was determined based on dose-limiting toxicities. Other assessments included safety, pharmacokinetic profile, and antitumour activity according to the Response Evaluation Criteria in Solid Tumors.

RESULTS

The study enrolled 21 patients. The maximum tolerated dose for BI 2536 was determined to be 60 mg for the study schedule. Dose-limiting toxicities included hematologic events, hypertension, elevated liver enzymes, and fatigue. The most frequently reported drug-related adverse events were mild-to-moderate fatigue, leukopenia, constipation, nausea, mucosal inflammation, anorexia, and alopecia. The pharmacokinetics of BI 2536 were linear within the dose range tested. Plasma concentration profiles exhibited multi-compartmental pharmacokinetic behaviour, with a terminal elimination half-life of 20-30 hours.

CONCLUSIONS

In the present study, BI 2536 showed an acceptable safety profile warranting further investigation of Plk1 inhibitors in this patient population.

Prediction of neutropenia-related effects of a new combination therapy with the anticancer drugs BI 2536 (a Plk1 inhibitor) and pemetrexed

This study investigated the feasibility of predicting the neutropenia-related effects of a therapy that combines the investigational drug BI 2536 (inhibitor of Polo-like kinase 1) and pemetrexed, an approved anticancer drug. Predictions were arrived at using the pharmacokinetic/pharmacodynamic (PK/PD) parameters of each of the drugs obtained from monotherapy studies and assuming that the neutropenic effect is additive when the drugs are administered as a combination therapy. Subsequently, a PK/PD model was developed to determine whether this assumption of additive effect was reasonable in relation to these two drugs. All analyses and simulations were performed using the population approach in NONMEM, version VI.

An open label phase II trial of the Plk1 inhibitor BI 2536, in patients with sensitive relapse small cell lung cancer (SCLC)

8108

Background: BI 2536 is a potent, selective inhibitor of polo-like kinase 1 (Plk1), a regulator of mitotic progression. BI 2536 demonstrated favorable tolerability and antitumor activity in phase I trials. We investigated the antitumor efficacy, safety and PK of BI 2536 in patients (pts) with sensitive relapse SCLC. Methods: This open label single arm phase II study followed a Gehan two-stage design. Primary objective was to determine the antitumor efficacy of BI 2536 in SCLC pts with disease recurrence ≥60 days after completion of first-line chemotherapy. 18 pts had to complete 2 courses to be evaluable for stage 1 analysis. In case of ≥2 partial or complete antitumor responses (RECIST criteria), stage 2 accrual would continue until 40 pts were entered. Patients received 200 mg BI 2536 as a 1h i.v. infusion on Day 1 every 3 weeks. Dose escalation to 250 mg (cycle 3 onwards) was encouraged in pts with <Grade 2 drug related non-hematologic and <Grade 3 hematologic toxicity. Results: 23 pts (14 female, 9 male, 21 extensive disease, 2 limited disease), median age 60 yrs (range: 35–77) were treated. All patients had disease recurrence >60 days after completion of first-line therapy. Of 23 pts, no objective antitumor responses were observed, 7 had stable disease as best response, 14 had progression, 2 were not evaluable. A median of 2 courses were given, up to a maximum of 12 in 1 pt. The PFS rate at 3 months was 25%. Due to the lack of antitumor responses, trial accrual was terminated after stage 1. Overall, BI 2536 was well tolerated. Frequent AEs were neutropenia (48%), fatigue (39%), nausea (30%), anemia, vomiting, constipation (26% each), and thrombocytopenia (22%). Drug related grade 3/4 AEs were neutropenia (13%/26%), grade 3/4 thrombocytopenia (1 pt each), grade 3/4 anemia (1 pt each), grade 4 sepsis (1 pt), Grade 4 ARDS (1 pt) and Grade 3 fatigue (1 pt). PK analyses indicate that BI 2536 has high clearance (>1,000 mL/min) and quickly distributes in multiple compartments in a large volume of distribution (>1,000 L). Estimated elimination half-life was >25 h. Conclusions: BI 2536 was well tolerated in relapsed SCLC pts, but demonstrated no convincing antitumor efficacy after stage I of the study. Therefore, BI 2536 will not be assessed further as a single agent in SCLC.

[Table: see text]

A phase I repeated dose escalation study of the Polo-like kinase 1 inhibitor BI 2536 in patients with advanced solid tumours

2038

Background: BI 2536 is a novel highly potent and selective inhibitor of the serine-threonine kinase polo-like kinase 1 (Plk1), which is a key regulator of cell cycle progression. Objectives of this trial were the assessment of the maximum tolerated dose (MTD), overall safety, pharmacokinetics and efficacy of BI 2536 given intravenously. Methods: Sequential cohorts of 3 to 6 patients (pts) with pretreated advanced or metastatic solid tumours received intravenous infusions of BI 2536 on days 1 and 8 of a 3-week treatment course following a toxicity guided dose escalation design. Further treatment courses were administered to pts in the absence of disease progression and if toxicity after a 3-week treatment course had resolved. Results: A total of 42 pts was treated at doses of 25 mg (n=3), 50 mg (n=3), 100 mg (n=22), 125 mg (n=5), 150 mg (n=6) and 200 mg (n=3). Reversible CTCAE grade ≥ 3 neutropenia in 14/42 pts represented the main drug related toxicity with an incidence of 3/5 in the 125 mg cohort, 4/6 in the 150 mg and 2/2 in the 200 mg dose cohorts. Dose limiting toxicity (DLT) was defined as drug related toxicity prohibiting administration of the day 8 dose of BI 2536 (hematologic: CTCAE ≥ 3 grade, non-hematologic toxicity: CTCAE ≥ 2). No DLT other than d8 neutropenia was observed. The MTD was defined at 100 mg for the given day 1 and 8 schedule. Further related adverse events (AE’s) were of mild to moderate intensity (CTCAE grade ≤ 2). There were no related AEs resulting in study discontinuation. Preliminary PK analysis showed dose proportionality of Cmax and AUC0-∞ with a high clearance (∼ 1500 mL/min) and a high volume of distribution (∼ 2000 L). No accumulation from d1 to d8 occurred. Patients were treated for up to 8 courses without evidence of accumulating toxicity. No objective responses were observed according to RECIST criteria in this heavily pretreated patient population. Conclusions: In summary BI 2536 is a Plk1 inhibitor with a favorable PK and safety profile at the tested dose and schedule. Neutropenia as a mechanism-related toxicity indicates target inhibition in vivo.

[Table: see text]

A phase I study of two administration schedules of the Polo-like kinase 1 inhibitor BI 2536 in patients with advanced solid tumors

3069

Background: BI 2536 is a novel highly potent and selective inhibitor of the serine-threonine Polo-like kinase 1 (Plk1), a key regulator of cell cycle progression. Objectives of this trial were the assessment of the maximum tolerated dose (MTD), overall safety, pharmacokinetics and preliminary efficacy of BI 2536 given intravenously. Methods: Sequential cohorts of 3–6 patients (pat) with advanced or metastatic solid tumors received infusions of BI 2536 following a toxicity guided dose escalation design. In the first part of the trial, a single administration was given every 21 days (d1). After completion of accrual BI 2536 was administered on three consecutive days (d1–3) in a dose intensified schedule in additional patient cohorts. Further treatment courses were given in the absence of disease progression and after recovery of toxicity after a 3-week observation period. Results: Of the 52 entered patients a total of 40 pts were treated at doses of 25 (n=3), 50 (n=3), 100 (n=3), 200 (n=25) and 250 mg (n=6) in the d1 schedule. The MTD for the d1 schedule was defined at 200 mg. Dose limiting toxicity (DLT) consisted of neutropenic infection and occurred in 2/6 patients at 250 mg. Reversible CTCAE grade ≥ 3 neutropenia represented the main drug related toxicity with an incidence of 13/25 in the 200 mg and 5/6 in the 250 mg dose cohorts. Alopecia was reported in 20% of pat. Further common drug related adverse events (AE’s) consisted of nausea (38%), anorexia (25%) fatigue (18%), vomiting (18%) and mucositis (12%) and were mostly of mild to moderate intensity (CTCAE grade ≤ 2). PK analysis showed dose proportionality of Cmax and AUC0-∞ with a high clearance (∼ 1400 mL/min) and a high volume of distribution (∼ 1200 L). A correlation between exposure and neutropenia was observed. Patients were treated for up to 10 courses without evidence of accumulating toxicity. One partial response was observed in a patient with metastatic squamous cell head and neck cancer treated at 250 mg. The DLT was determined at 3x70 mg in the dose intensified schedule with the MTD still pending. Conclusion: In summary BI 2536 is a Plk1 inhibitor with a favorable safety and PK profile. Neutropenia as a mechanism-related toxicity indicates target inhibition in vivo. First signs of antitumor activity were observed.

[Table: see text]

Rhenium 188-labeled anti-CD66 (a, b, c, e) monoclonal antibody to intensify the conditioning regimen prior to stem cell transplantation for patients with high-risk acute myeloid leukemia or myelodysplastic syndrome: results of a phase I-II study

The conditioning regimen prior to stem cell transplantation in 36 patients with high-risk acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) was intensified by treating patients with a rhenium 188-labeled anti-CD66 monoclonal antibody. Dosimetry was performed prior to therapy, and a favorable dosimetry was observed in all cases. Radioimmunotherapy with the labeled antibody provided a mean of 15.3 Gy of additional radiation to the marrow; the kidney was the normal organ receiving the highest dose of supplemental radiation (mean 7.4 Gy). Radioimmunotherapy was followed by standard full-dose conditioning with total body irradiation (12 Gy) or busulfan and high-dose cyclophosphamide with or without thiotepa. Patients subsequently received a T-cell-depleted allogeneic graft from a HLA-identical family donor (n = 15) or an alternative donor (n = 17). In 4 patients without an allogeneic donor, an unmanipulated autologous graft was used. Infusion-related toxicity due to the labeled antibody was minimal, and no increase in treatment-related mortality due to the radioimmunoconjugate was observed. Day +30 and day +100 mortalities were 3% and 6%, respectively, and after a median follow-up of 18 months treatment-related mortality was 22%. Late renal toxicity was observed in 17% of patients. The relapse rate of 15 patients undergoing transplantation in first CR (complete remission) or second CR was 20%; 21 patients not in remission at the time of transplantation had a 30% relapse rate. (Blood. 2001;98:565-572)

Normal structure of NFKB2, C-REL and BCL-3 gene loci in lymphoproliferative and myeloproliferative disorders

NF-kappaB/rel transcription factors are crucial regulators of development, differentiation and apoptosis of both lymphoid and myeloid lineages. There is increasing evidence for an involvement of NF-kappaB/rel proteins in lymphomagenesis and resistance of lymphoid tumors to the induction of apoptosis. Structural alterations of the NF-kappaB/rel genes NFkappaB2, c-rel and bcl-3 have been shown to result in increased NF-kappaB/rel activity. Because we observed strong constitutive NF-kappaB/rel binding activity in chronic lymphocytic leukemia of the B-cell type (B-CLL) which may contribute to resistance against cytotoxic drugs we studied the genomic organisation of NFkappaB2, c-rel and bcl-3 gene loci in a panel of lymphoproliferative disorders (n=81) with an emphasis on B-CLL (n=47). The method of genomic Southern blotting using cDNAs of the respective genes was used. In spite of the role of NF-kappaB/rel in myeloid maturation there is no data available as to the occurrence of NF-kappaB/rel rearrangements in chronic myeloproliferative syndromes (cMPS). For this reason we included a small panel of cMPS patients (n=16). Southern Blotting revealed a germline configuration of NFkappaB2, c-rel and bcl-3 loci in all NHL and cMPS patients examined. Our results demonstrate that structural alterations of NFkappaB2, c-rel and bcl-3 genes at the Southern Blotting level are rare events that do not contribute to lymphoid or myeloid transformation in the majority of NHL or cMPS patients.

The SH2-containing adapter protein GRB10 interacts with BCR-ABL

Bcr-Abl is an oncogenic tyrosine kinase expressed in tumor cells of CML and a subset of ALL which in its unregulated and activated state is thought to cause cell transformation and leukemia. Bcr-Abl contains several autophosphorylation sites which serve as potential docking sites for SH2-containing signaling molecules. Mutational analysis has indicated that these autophosphorylation sites play a critical role in the transforming capability of Bcr-Abl. It has been shown that the SH2-containing adapter protein Grb2 binds to the autophosphorylation site Tyr(p)177 whereby it couples Bcr-Abl to the Ras pathway. The biological consequences of this interaction, however, are presently unclear. A Tyr177-mutated Bcr-Abl which lacks the ability to interact with the Grb2-SH2 domain still transforms myeloid cells and generates tumors in nude mice. We performed a yeast two-hybrid screen to identify signaling proteins which bind to distinct Bcr-Abl autophosphorylation sites. Autophosphorylation of Bcr-Abl in yeast was accomplished by using the DNA binding protein LexA which permits dimerization and crossphosphorylation of the fused bait. Using a LexA-Bcr-Abl full length fusion protein as bait, we identified several SH2-containing proteins. Among them we confirmed molecules already shown by others to interact with Bcr-Abl, in vivo, including Grb2, PI-3-kinase and Crk indicating that dimerization in yeast leads to autophosphorylation of tyrosine residues crucial for Bcr-Abl signaling in vivo. More importantly, we identified the SH2-containing protein Grb10 as a new binding partner for Bcr-Abl. This binding occurs in a phosphotyrosine-dependent manner at Bcr sites of Bcr-Abl. Both Abl and Bcr alone, as well as a kinase-defective Bcr-Abl, failed to interact with Grb10 in yeast. Mutational analysis uncovered a new SH2 binding site in Bcr-Abl located between Bcr aa242-446, which is different from the Grb2 binding site. Binding could be demonstrated in vitro and also in vivo as shown by co-immunoprecipitation analysis in CML cells. Using a temperature sensitive Bcr-Abl stably overexpressed in hematopoetic cells, we demonstrated that complex formation of Grb10 with Bcr-Abl was kinase activation-dependent in vivo. Notably, a Bcr-Abl mutant protein (Bcr/1-242-Abl) which lacks the ability to interact with Grb10 partially alleviated IL-3 dependence of Ba/F3 cells, indicating that the Grb10/Bcr-Abl interaction is important for Bcr-Abl-induced IL-3 independence of Ba/F3 cells. In addition, the Bcr/1-242-Abl mutant has a reduced capacity to induce focus formation in fibroblasts.

Concomitant manifestation of systemic lupus erythematosus and low-grade non-Hodgkin's lymphoma

We describe a patient presenting with systemic lupus erythematosus (SLE) and concomitant low-grade (Ig) non-Hodgkin's lymphoma of the B cell type (B-NHL). Although the association of autoimmune disorder and lymphoma is well conceived, there is only scarce information available as to the simultaneous occurrence of both disease conditions in one patient. As in this patient diagnosis of Ig B-NHL was also based on the detection of a monoclonal population of CD5+ B lymphocytes, and given that the polyclonal expansion of CD5+ B cells has been previously reported in rheumatoid arthritis (RA), Sjogren's syndrome (SS) and single cases of SLE, the observations we made in this patient led us to discuss the role of the CD5+ population in the development of rheumatic disorders and concomitant lymphoid malignancy. Moreover, since impaired production rates of interleukin 3 (IL-3) and interleukin 4 (IL-4) have been associated with an abnormal expansion of CD5, lymphoma cells and seeing that soluble interleukin 2 receptor (sIL-2R) serum levels were found to be positively correlated with disease activity both in SLE and Ig B-NHL, these parameters were investigated and related to the patient's disease state throughout the entire clinical observation period.

Construction of a novel bifunctional biogenic amine receptor by two point mutations of the H2-histamine receptor

BACKGROUND

H2-histamine receptors mediate a wide range of physiological functions extending from stimulation of gastric acid secretion to induction of human promyelocyte differentiation. We have previously cloned the H2-histamine receptor gene and noted that only three amino acids on the receptor were sufficient to define its specificity and selectivity. Despite only modest overall amino acid homology (34% amino acid identity and 57.5% similarity) between the H2-histamine receptor and the receptor for another monoamine, the beta 2-adrenergic receptor, there is remarkable similarity at their critical ligand binding sites. We hypothesized that, if the specificity and selectivity of both receptors are invested in just three amino acids, it should be possible to convert one of the receptors into one that recognizes the ligand of the other by simple mutations at only one or two sites.

MATERIAL AND METHODS

We explored the effect of two single mutations in the fifth transmembrane domain of the H2-histamine receptor, which encompasses the sites that determine H2 selectivity. The canine H2 receptor gene was mutated at Asp186 and Gly187 (Asp186 to Ala186 and Gly187 to Ser187) by oligonuceotide directed mutagenesis. The coding region of both the wild-type and mutated H2 receptors was subcloned into the eukaryotic expression vector, CMVneo, and stably transfected into Hepa cells and L cells. The biological activity of histamine and epinephrine on the expressed receptor was examined by measurement of cellular cAMP production and inositol trisphosphate formation.

RESULTS

Hepa cells transfected with the Ala186-Ser187 mutant H2 receptor demonstrated a biphasic rise in cAMP in response to epinephrine with an early phase (ED50 approximately 10(-11) M) that could be inhibited by both propranolol and cimetidine. Epinephrine also induced IP3 generation in the same cells, a biological response that is characteristic of activation of the wild-type H2 but not of the beta-adrenergic receptor. L cells transfected with the Ala186-Ser187 mutant H2 receptor also responded to epinephrine in a cimetidine and propranolol inhibitable manner.

CONCLUSIONS

We converted the H2-histamine receptor into a bifunctional one that has characteristics of both histamine and adrenergic receptors by two simple mutations. These results support the hypothesis that ligand specificity is determined by only a few key points on a receptor regardless of the structure of the remainder of the molecule. Our studies have important implications on the design of pharmacological agents targeted for action at physiological receptors.

Molecular cloning of a novel melanocortin receptor

Using the technique of the polymerase chain reaction primed with oligonucleotides based on the homologous transmembrane regions of seven transmembrane G protein-linked receptors, we isolated three full-length human genes that encode a novel subgroup of this receptor family. Recently, two of these receptors were identified as specific for alpha-melanocyte-stimulating hormone (alpha-MSH) and adrenocorticotropic hormone. We report the molecular cloning and pharmacologic characterization of a third member of this subgroup. The gene for this receptor encodes a protein of 361 amino acids in length. Its pharmacology characterizes it as an MSH receptor specific to the heptapeptide core common to adrenocorticotropic hormone and alpha-, beta-, and gamma-MSH. By Northern blot hybridization and polymerase chain reaction, it is expressed in brain, placental, and gut tissues but not in melanoma cells or in the adrenal gland. These findings may yield insight into the physiology of peptides derived from pro-opiomelanocortin post-translational processing.

Molecular basis for the interaction of histamine with the histamine H2 receptor

We undertook these studies to characterize the molecular basis of the interaction of histamine with the H2 receptor. Key areas of homology in the structures of the histamine H2 and beta 2 adrenergic receptor suggested specific transmembrane amino acids that might be important for binding of histamine. A third transmembrane aspartic acid of the histamine receptor (Asp98), thought to serve as a counter anion that interacts with the cationic amine moiety of histamine, was mutated to Asn98, and the mutated receptor was expressed in Hepa cells. Removal of the negatively charged amino acid abolished both binding of the H2 receptor antagonist [methyl-3H]tiotidine and histamine stimulated increases in cellular cAMP content. Mutation of a fifth transmembrane aspartic acid (Asp186) to Ala186 or Asn186 by itself or in conjunction with mutation of another fifth transmembrane amino acid (Thr190 to Ala190) resulted in a loss of [methyl-3H] tiotidine binding, although the generation of cAMP in response to histamine was maintained. The histamine receptor with only a Thr190 to Ala190 or Cys190 mutation retained the ability to bind [methyl-3H]tiotidine, but both the affinity and efficacy of binding were reduced. These data lead us to propose a model for histamine binding in which Asp98 is essential for histamine binding and action, Asp186 defines H2 selectivity, and Thr190 is important in establishing the kinetics of histamine binding, but is not essential for H2 selectivity.

Molecular cloning of the human histamine H2 receptor

We utilized the technique of polymerase chain reaction with oligonucleotide primers based upon the nucleotide sequence of the canine H2 histamine receptor gene which we recently isolated to clone its human homologue. Transfection of a construct of this gene in Colo-320 DM cells led to the expression of a receptor that bound to [methyl-3H] tiotidine and was linked to 3',5'cyclic adenosine monophosphate (cAMP) generation in response to histamine. Both cAMP generation and [methyl-3H] tiotidine binding were inhibited with the H2 histamine receptor selective antagonist cimetidine but not diphenhydramine or thioperamide which are, respectively, H1 and H3 histamine receptor antagonists. These data confirm that we have successfully cloned a novel gene encoding the human H2 histamine receptor.

Vagally induced release of gastrin, somatostatin and bombesin-like immunoreactivity from perfused rat stomach. Effect of stimulation frequency and cholinergic mechanisms

The isolated stomach of rats was vascularly perfused to measure the secretion of gastrin, somatostatin (SLI) and bombesin-like immunoreactivity (BLI). The gastric lumen was perfused with saline pH 7 or pH 2, and electrical vagal stimulation was performed with 1 ms, 10 V and 2, 5 or 10 Hz, respectively. Atropine was added in concentrations of 10(-9) or 10(-7) M to evaluate the role of cholinergic mechanisms. In control experiments, vagal stimulation during luminal pH 2 elicited a significant increase of BLI secretion only at 10 Hz but not at 2 and 5 Hz. Somatostatin release was inhibited independent of the stimulation frequency employed. Gastrin secretion at 2 Hz was twice the secretion rates observed at 5 and 10 Hz, respectively. At luminal pH 7 BLI rose significantly at 5 and 10 Hz. SLI secretion was decreased by all frequencies. Gastrin secretion at 2 and 5 Hz was twice as high as during stimulation with 10 Hz. Atropine at doses of 10(-9), 10(-8), 10(-7) and 10(-6) M had no effect on basal secretion of BLI, SLI and gastrin. At luminal pH 2, atropine increased dose-dependently the BLI response at 2 and 5 but not at 10 Hz. The decrease of SLI during 2 and 5 Hz but not 10 Hz was abolished by atropine 10(-9) M. SLI was reversed to stimulation during atropine 10(-7) M at all frequencies. The rise of gastrin at 2 Hz was reduced by 50%. At luminal pH 7, atropine had comparable effects with a few differences: the BLI response at 10 Hz was augmented and the gastrin response to 2 and 5 Hz was reduced. In conclusion the present data demonstrate a frequency and pH-dependent stimulation of BLI and gastrin release. The stimulation of BLI is predominantly due to atropine-insensitive mechanisms while muscarinic cholinergic mechanisms exert an inhibitory effect on BLI release during lower stimulation frequencies (2 and 5 Hz) independent of the intragastric pH and also during higher frequencies at neutral pH. Both, atropine sensitive and insensitive mechanisms are activated frequency dependent. The atropine-sensitive cholinergic mechanisms but not the noncholinergic mechanisms involved in regulation of G-cell function are pH and frequency dependent. Somatostatin is regulated largely independent of stimulation frequency and pH by at least two pathways involving cholinergic mechanisms of different sensitivity to atropine. These data suggest a highly differentiated regulation of BLI, gastrin and SLI secretion and the interaction between these systems awaits further elucidation.