scholarly journals MON-LB57 Impact of Imputation Method and Response Cutoffs on Results From the Phase 3 OPTIMAL Study of Oral Octreotide Capsules in Adult Patients With Acromegaly

2020 ◽  
Vol 4 (Supplement_1) ◽  
Author(s):  
Susan Leanne Samson ◽  
Lisa B Nachtigall ◽  
Maria Fleseriu ◽  
Murray B Gordon ◽  
Mojca Jensterle ◽  
...  
Keyword(s):  
Primary Endpoint ◽  
Response Rate ◽  
Clinical Symptoms ◽  
Somatostatin Receptor ◽  
Study Drug ◽  
Sensitivity Analyses ◽  
Biochemical Response ◽  
Phase 3 ◽  
Somatostatin Receptor Ligands ◽  
Igf I

Abstract Objective: The phase 3 CHIASMA OPTIMAL study assessed efficacy and safety of oral octreotide capsules (OOC) in patients with acromegaly controlled on injectable somatostatin receptor ligands (SRLs). Sensitivity analyses were conducted for efficacy endpoints using two methods of imputation (i.e., the process of replacing clinical data with substitution values) to address missing data points due to some subjects reverting back to their prior injectable SRL treatment. Methods: Patients were ≥18 years of age and had evidence of active acromegaly with an average IGF-I ≤ 1.0 x ULN (utilizing the IDS iSYS assay calibrated to WHO recombinant reference standard 02/254). At baseline, patients were randomized to receive OOC or placebo for 36 weeks. The primary endpoint was proportion of patients maintaining biochemical response, defined as IGF-I ≤1.0 x ULN (2-value average at weeks 34 and 36) (Samson et al. ENDO 2020). Per study protocol, patient study discontinuations were considered non-responders regardless of clinical response at the time of discontinuation (non-response imputation). Additional exploratory analyses were performed utilizing the last observation carried forward (LOCF) analysis, as well as a completers analysis of response among the subgroup that completed the entire 36 weeks on study drug. The response rates reported for the primary end point are slightly adjusted for stratification differences as prespecified in the statistical analysis plan. Results: Twenty-eight patients received OOC and 12 failed to maintain biochemical response based on the primary endpoint. Seven of these 12 patients discontinued treatment early - 5 due to treatment failure and 2 due to AEs. The remaining 5 patients completed the 36-week protocol on study drug. Of these 5 patients, 4 had IGF-I values between >1.0 and ≤1.3 x ULN and 1 completed the study with an IGF-I of 1.7 x ULN with no clinical symptoms. 58.2% of patients in the OOC group met the primary endpoint of maintenance of biochemical response at the end of study using the non-response imputation. Using LOCF imputation, 64.3% (18/28) of patients met this endpoint. Of those completing the study (N=21), 76.2% maintained response. Conclusion: CHIASMA OPTIMAL primary endpoint was assessed using the non-response imputation for patients who discontinued treatment early, with a 58.2% response rate. However, when assessing the response rate based on LOCF imputation, or in study completers, similar to other phase 3 studies for acromegaly, the rate was imputed at 64.3% and 76.2%, respectively.

scholarly journals A Phase 3 Large International Noninferiority Trial (MPOWERED): Assessing Maintenance of Response to Oral Octreotide Capsules in Comparison to Injectable Somatostatin Receptor Ligands

2021 ◽  
Vol 5 (Supplement_1) ◽  
pp. A517-A517
Author(s):  
Maria Fleseriu ◽  
Alexander V Dreval ◽  
Yulia Pokramovich ◽  
Irina Bondar ◽  
Elena Isaeva ◽  
...  
Keyword(s):  
Somatostatin Receptor ◽  
Safety Data ◽  
Biochemical Response ◽  
Receptor Ligands ◽  
Phase 3 ◽  
Landmark Analysis ◽  
End Point ◽  
Noninferiority Trial ◽  
Somatostatin Receptor Ligands ◽  
Igf I

Abstract Background: MPOWERED, a large phase 3 trial, assessed maintenance of response to oral octreotide capsules (OOC; MYCAPSSA®) compared to injectable somatostatin receptor ligands (iSRLs) in patients with acromegaly who responded to OOC and iSRLs (octreotide or lanreotide). OOC were recently approved in the US for patients with acromegaly who responded to and tolerated iSRLs. Methods: Eligibility criteria included age 18-75 years at screening, acromegaly diagnosis, disease evidence, biochemical control (insulin-like growth factor I [IGF-I] <1.3 × upper limit of normal [ULN] and mean integrated growth hormone [GH] <2.5 ng/mL) at screening, and ≥6 months’ iSRL treatment. Effective OOC dose was determined in a 26-week Run-in phase. Eligible patients (IGF-I <1.3 × ULN and mean integrated GH <2.5 ng/mL, week 24) were randomized to a 36-week controlled treatment phase (RCT), receiving OOC or iSRLs starting at week 26. The primary end point was a noninferiority assessment of proportion of patients biochemically controlled in the RCT (IGF-I <1.3 × ULN using time-weighted average). Other end points included nonresponse imputation of the primary end point, landmark analysis using proportion of responders based on average of last 2 IGF-I values at end of RCT, and change from baseline RCT (week 26) IGF-I and GH levels. Results: Of 146 enrolled patients, 92 entered the RCT (OOC, n=55; iSRLs, n=37). Both arms were well balanced for age, sex, and acromegaly duration. OOC demonstrated noninferiority to iSRLs in maintaining biochemical response, with 91% (CI, 80%-97%) of OOC and 100% (CI, 91%-100%) of iSRL groups maintaining control during the RCT. Of those responding at end of Run-in, 96% of patients on OOC maintained response during RCT. Using nonresponse imputation, 89% of OOC and 95% of iSRL groups were biochemically controlled in RCT. Landmark analysis of those respnding at end of Run-in showed that 94% of patients in each group maintained response at RCT end. In both groups, IGF-I levels were stable in the RCT, average IGF-I at baseline and RCT end being 0.9 × ULN (OOC) and 0.8 × ULN (iSRL). Mean change in GH from RCT start to RCT end was -0.03 ng/mL (OOC) and +0.29 ng/mL (iSRL). Safety data were mostly similar between groups; the OOC group did not experience injection site reactions. Conclusion: In this noninferiority trial in patients with acromegaly, OOC demonstrated maintenance of biochemical response compared to iSRLs. Results support the efficacy of OOC as a possible iSRL alternative.

scholarly journals MON-LB53 Prior Injectable Somatostatin Receptor Ligand Dose Does Not Predict Oral Octreotide Response In The Treatment Of Acromegaly: Results From The Phase 3 OPTIMAL Study

2020 ◽  
Vol 4 (Supplement_1) ◽  
Author(s):  
Lisa B Nachtigall ◽  
Susan Leanne Samson ◽  
Maria Fleseriu ◽  
Marek Bolanowski ◽  
William Henry Ludlam ◽  
...  
Keyword(s):  
Treatment Effect ◽  
Low Dose ◽  
Somatostatin Receptor ◽  
Prior Treatment ◽  
High Dose ◽  
Low Doses ◽  
Phase 3 ◽  
Somatostatin Receptor Ligands ◽  
Igf I ◽  
High Doses

Abstract Background: Injectable somatostatin receptor ligands (SRLs) are the most widely used therapy to control acromegaly. Oral octreotide capsules have been formulated as a potential therapy for this disorder and the efficacy and safety was evaluated in the CHIASMA OPTIMAL prospective phase 3 study in patients with acromegaly who were controlled on injectable SRL treatment of varying doses (Samson et al. ENDO 2020). Methods: Patients with confirmed acromegaly, who had been receiving a stable dose of injectable SRL (≥3 months) up until study entry, were randomized to receive octreotide capsules (40 mg/day) or placebo for 36 weeks. Patients were dose titrated to 60 or 80 mg of oral octreotide or equivalent placebo through week 24 at the investigator’s discretion based on increase of IGF-I levels or worsening of acromegaly signs and symptoms. The primary efficacy endpoint was the proportion of patients who maintained their biochemical response at the end of 36 weeks, defined as average IGF-I ≤1 x ULN between Weeks 34 and 36. An analysis evaluated maintenance of response based on prior dose of injectable SRL. Prior doses of injectable SRL were categorized based on the following classifications: octreotide 10 mg every 4 weeks or lanreotide 60 mg every 4 weeks or 120 mg every 8 weeks were stratified as low; octreotide 20 mg every 4 weeks or lanreotide 90 mg every 4 weeks or 120 mg every 6 weeks were stratified as medium; octreotide 30 mg or 40 mg or lanreotide 120 mg every 4 weeks were stratified as high. Randomization was stratified based on low dose vs med/high dose and efficacy results compared for these strata. The response rates reported for the primary end point are slightly adjusted for stratification differences as prespecified in the statistical analysis plan. Results: Six patients (21.4%) in the octreotide capsule group had received prior treatment with low doses of injectable SRLs while 22 (78.6%) had received prior treatment with medium-high doses of injectable SRLs. Maintenance of response was observed in 16 patients receiving oral octreotide. This included 66.7% of patients (n=4) previously receiving low doses of injected SRLs and 54.5% of patients (n=12) on medium-high injected doses. The treatment effect was consistent irrespective of prior dose of injectable SRL (odds ratio: 5.4 in low dose and 5.9 in medium-high dose). Conclusion: The CHIASMA OPTIMAL study recruited a population receiving predominantly medium-high doses of injectable SRLs and demonstrated maintenance of response in 58% of patients. Oral octreotide treatment effect was consistent irrespective of prior dose of injectable SRL.

scholarly journals Addition of Cabergoline to Oral Octreotide Capsules May Improve Biochemical Control in Patients With Acromegaly Who Are Inadequately Controlled With Monotherapy

2021 ◽  
Vol 5 (Supplement_1) ◽  
pp. A518-A519
Author(s):  
Maria Fleseriu ◽  
Akexander V Dreval ◽  
Yulia Pokramovich ◽  
Irina Bondar ◽  
Elena Isaeva ◽  
...  
Keyword(s):  
Somatostatin Receptor ◽  
Combined Treatment ◽  
Study Drug ◽  
Receptor Ligands ◽  
Biochemical Control ◽  
Somatostatin Receptor Ligands ◽  
Controlled Treatment ◽  
The Us ◽  
Igf I ◽  
First Time

Abstract Background: Oral octreotide capsules (OOC; MYCAPSSA®) are approved in the US for individuals with acromegaly who responded to and tolerated treatment with injectable somatostatin receptor ligands (iSRLs). Add-on cabergoline therapy has shown effectiveness in patients previously inadequately controlled with iSRLS.1 The phase 3 MPOWERED trial assessed maintenance of response with OOC compared to iSRLs. Patients receiving OOC and ineligible for randomized controlled treatment (RCT) phase were eligible for a sub-study evaluating combination therapy with cabergoline, a dopamine agonist. Methods: Patients who fail to respond to 80 mg/d OOC for ≥2 weeks during the 26-week Run-in phase, or ineligible to enter the RCT on 80 mg/d OOC, due to inadequate biochemical control (insulin-like growth factor I [IGF-I] ≥1.3 × upper limit of normal [ULN] to <2 × ULN or IGF-I <1.3 × ULN and mean integrated growth hormone [GH] ≥2.5 ng/mL) were eligible for sub-study combination OOC 80 mg/d and cabergoline ≤3.5 mg/wk (fixed algorithm) for 36 weeks. End points included categorical changes in IGF-I and mean GH levels at sub-study end and adverse event (AE) incidence and severity. Echocardiogram was performed at sub-study start and every 12 weeks after. Results: Of 146 patients enrolled in MPOWERED, 14 entered the combination sub-study, 9 having IGF-I ≥1.3 × ULN at sub-study start. Final cabergoline doses were 1 (n=5), 2 (n=3), 3 (n=1), and 3.5 mg (n=5) with 25.4-week (SD, 14.1) mean treatment duration. Week 36 IGF-I improved in most patients (n=12; 85.7%). Of 9 patients with IGF-I ≥1.3 × ULN at sub-study start, 5 (55.6%; 95% CI, 21.2%-86.3%) exhibited IGF-I decreased to predefined responder range (<1.3 × ULN) by week 36. AE incidence and nature with combined treatment were similar to known octreotide safety profile and acromegaly disease burden. There were no serious AEs or AEs leading to discontinuation of either sub-study drug. Conclusion: We have shown for the first time the benefit of an all-oral combination treatment for acromegaly and avoidance of injection-related burdens. Addition of cabergoline to OOC yielded biochemical control improvement (IGF-I reduction) in patients inadequately controlled with OOC monotherapy. As both combination and OOC monotherapy safety profiles were similar, adjunctive cabergoline may be helpful in patients with acromegaly who do not achieve adequate biochemical control on OOC alone. 1Giustina A, et al. Nat Rev Endocrinol. 2014;10(4):243-248.

scholarly journals Maintenance of Acromegaly Control in Patients Switching From Injectable Somatostatin Receptor Ligands to Oral Octreotide

2020 ◽  
Vol 105 (10) ◽  
pp. e3785-e3797 ◽  
Author(s):  
Susan L Samson ◽  
Lisa B Nachtigall ◽  
Maria Fleseriu ◽  
Murray B Gordon ◽  
Marek Bolanowski ◽  
...  
Keyword(s):  
Somatostatin Receptor ◽  
Control Measures ◽  
Double Blind Study ◽  
Efficacy And Safety ◽  
Receptor Ligands ◽  
Biochemical Control ◽  
Phase 3 ◽  
Double Blind ◽  
Somatostatin Receptor Ligands ◽  
End Of Treatment

Abstract Purpose The phase 3 CHIASMA OPTIMAL trial (NCT03252353) evaluated efficacy and safety of oral octreotide capsules (OOCs) in patients with acromegaly who previously demonstrated biochemical control while receiving injectable somatostatin receptor ligands (SRLs). Methods In this double-blind study, patients (N = 56) stratified by prior SRL dose were randomly assigned 1:1 to OOC or placebo for 36 weeks. The primary end point was maintenance of biochemical control at the end of treatment (mean insulin-like growth factor 1 [IGF-1] ≤ 1.0 × upper limit of normal [ULN]; weeks 34 and 36). Time to loss of IGF-1 response and proportion requiring reversion to injectable SRLs were assessed as broader control measures. Results Mean IGF-1 measurements were 0.80 and 0.97 × ULN for OOC and 0.84 and 1.69 × ULN for placebo, at baseline and end of treatment, respectively. Mean growth hormone (GH) changed from 0.66 to 0.60 ng/mL for OOCs and 0.90 to 2.57 ng/mL for placebo. Normalization of IGF-1 levels (≤ 1.0 × ULN) was maintained in 58.2% for OOCs vs 19.4% for placebo (P = .008); GH levels were maintained (< 2.5 ng/mL) in 77.7% for OOC vs 30.4% for placebo (P = .0007). Median time to loss of response (IGF-1 > 1.0 or ≥ 1.3 × ULN definitions) for patients receiving placebo was 16 weeks; for patients receiving OOCs, it was not reached for both definitions during the 36-week trial (P < .0001). Of the patients in the OOC group, 75% completed the trial on oral therapy. The OOC safety profile was consistent with previous SRL experience. Conclusions OOCs may be an effective therapy for patients with acromegaly who previously were treated with injectable SRLs.

Phase II trial of BNC105P as second-line chemotherapy for advanced malignant pleural mesothelioma (MPM): Australasian Lung Cancer Trials Group and NHMRC Clinical Trials Centre Collaboration.

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. 7079-7079
Author(s):  
Anna K. Nowak ◽  
Chris Brown ◽  
Michael Millward ◽  
Brett Gordon Maxwell Hughes ◽  
David C. Bibby ◽  
...  
Keyword(s):  
Primary Endpoint ◽  
Response Rate ◽  
Single Agent ◽  
Objective Response ◽  
Tumour Response ◽  
Dose Intensity ◽  
Study Drug ◽  
Progression Free Survival ◽  
Second Line ◽  
Vascular Disrupting Agent

7079^ Background: BNC105P is a tubulin polymerization inhibitor that acts as a Vascular Disrupting Agent (VDA), has direct cytotoxic effects, and had preclinical and phase I activity in MPM. This aim of this study was to determine activity and safety of BNC105P as second-line therapy after pemetrexed and a platin in MPM. Methods: Eligible patients had progressive MPM, prior pemetrexed and platinum, measurable disease by modified RECIST, ECOG 0-1, and adequate organ and cardiovascular function. Important exclusions included recent thromboembolic, cardiovascular or cerebrovascular disease, or therapeutic anticoagulation. Pts received BNC105P (16 mg/m2 IV) Day 1 + 8 q21d until progression or toxicity. The primary endpoint was centrally reviewed objective tumour response rate (RR); the Simon 2-stage design assumed a RR of interest of 20% and a RR of no interest of 5%, with α = β = 0.05. Continuation past first stage accrual required >1 objective response in 24 patients. Results: 30 subjects were accrued over 10 months (90% male; median age 65 (range 41-83); 77% ECOG PS 1; histology epithelioid (67%), biphasic (10%), sarcomatoid (7%), other/unspecified (17%)). All pts received at least one dose of study drug; pts received a median of 2 cycles and median dose intensity was 100%. No significant haematologic, biochemical, peripheral neurotoxic or cardiac adverse events (AEs) including hypertension were observed. Grade 3 or 4 AEs occurred in 10 pts (33%). There were 2 deaths on study: 1 due to stroke, the other due to pneumonia and respiratory failure. We observed 1 partial response (3%) and 13 pts with SD as their best response (43%). Median progression free survival was 1.5 mo (95% CI 1.4-2.4); median overall survival was 8.7 mo (95% CI 3.8-NR). Lung function and QOL data was collected. Biomarker analyses correlating to pharmacological changes induced by BNC105P are ongoing and will be presented. Conclusions: BNC105P was safe and tolerable but its single agent response rate failed to meet the pre-specified primary endpoint of interest.

scholarly journals MON-320 Inter-Rater Reliability of T2 MRI Intensity of Somatotroph Adenomas; Endocrinologists vs. Neuroradiologist Pilot Study

2020 ◽  
Vol 4 (Supplement_1) ◽  
Author(s):  
Elena V Varlamov ◽  
José Miguel Hinojosa-Amaya ◽  
Dawn Lim Shao Ting ◽  
Maria Fleseriu ◽  
Joao Prola
Keyword(s):  
White Matter ◽  
Pilot Study ◽  
Medical Therapy ◽  
Somatostatin Receptor ◽  
Biochemical Response ◽  
Receptor Ligands ◽  
Rater Reliability ◽  
Post Surgery ◽  
Somatostatin Receptor Ligands ◽  
T2 Mri

Abstract Background MRI T2 hypointensity of growth hormone (GH) secreting pituitary adenomas (PA) has been associated with better biochemical response to somatostatin receptor ligands and has been suggested to be useful in selecting patients with expected favorable response for pre- and post-surgery medical therapy. However, in most imaging centers, T2 intensity measurement is not part of standard neuroradiologist (NR) reporting. Objective To assess whether endocrinologists (Es) can reliably measure PA T2 signal intensity by calculating inter-rater reliability between Es and NR. Methods Retrospective review of MRI in 20 patients with pituitary somatotroph macroadenoma randomly selected from an IRB-approved PA database who had preoperative MRI available. T2 MRI intensity of the solid portion of the PA was compared to the temporal gray matter (GM) and white matter (WM): hypo- (PA< WM), hyper- (PA> GM), and isointense (WM <PA <GM). Measurements were performed separately by a NR and by two Es trained to take measurements by the same NR. Statistics: SPSS 25; Cohen kappa (κ). Results Patient mean age was 47 ± 20 years, with 12 females; mean largest PA diameter was 22.6 mm (range 11-45 mm). NR measured 12 hyper-, 7 iso- and 1 hypo-intense PA. Agreement was moderate between NR and E#1 (κ 0.72, 95%CI 0.751-1.0, p<0.001) and NR and E#2 (κ 0.638, 95%CI 0.351-0976, p<0.001) and strong between E#1 and E#2 (κ=0.90, 95%CI 0.309-0.903, p=0.001). Hypointense PA (by NR) was read by both Es as isointense. One hyperintense PA (by NR) was read by both Es as isointense. One isointense PA was read by E#2 as hypointense. Overall adenomas were; 9 densely granulated GH, 5 sparsely granulated GH, 3 mixed GH and prolactin, 1 plurihormonal, 1 not classified, and 1 no surgical intervention. Discussion Inter-rater reliability between the 2 Es was strong, however, it was moderate between each E and the NR. Factors that likely contributed to difference in measurement are heterogeneity of the PA, MRI quality, selection bias in choosing “most appropriate” site to measure intensity of adenoma, gray and white matter. Es could be trained to interpret the T2 intensity, although reliability with NR is only moderate. Interestingly, in this sample majority of T2 PA were hyperintense, but densely granulated, suggesting that preoperative identification of densely granulated tumors, which are also predictive of favorable SRL response, might be limited. More studies are needed to assess T2 correlation with pathology. Conclusion As T2 intensity (hyper-, hypo- or iso-) on MRI might be predictive of biochemical response to medical therapy in some patients with PA, we recommend T2 intensity to be part of neuroradiology reporting protocol. Our pilot study showed that endocrinologists could read MRIs after adequate training, but there is only moderate correlation with neuroradiologists.

scholarly journals Oral Octreotide Capsules Lowered Incidence and Improved Severity of Acromegaly Symptoms Compared to Injectable Somatostatin Receptor Ligands—Results From the MPOWERED Trial

2021 ◽  
Vol 5 (Supplement_1) ◽  
pp. A522-A523
Author(s):  
Nienke Biermasz ◽  
Maria Fleseriu ◽  
Akexander V Dreval ◽  
Yulia Pokramovich ◽  
Irina Bondar ◽  
...  
Keyword(s):  
Somatostatin Receptor ◽  
Symptom Burden ◽  
Receptor Ligands ◽  
Phase 3 ◽  
Eligibility Criteria ◽  
Somatostatin Receptor Ligands ◽  
Severity Scores ◽  
The Us ◽  
Normal Mean ◽  
Index Of Severity

Abstract Background: Patients with acromegaly may have high symptom burden. The phase 3 MPOWERED trial assessed control of acromegaly by oral octreotide capsules (OOC; MYCAPSSA®) in comparison to injectable somatostatin receptor ligands (iSRLs) in patients responding to both OOC and iSRLs. iSRLs have been first-line medical treatment for patients with acromegaly for decades. OOC are newly approved in the US for patients previously controlled on iSRLs. Methods: Eligibility criteria for MPOWERED included acromegaly diagnosis, biochemical control of acromegaly (insulin-like growth factor I <1.3 × upper limit of normal; mean integrated growth hormone, <2.5 ng/mL) and ≥6 months’ iSRL (octreotide, lanreotide) treatment. Eligible patients entered a 26-week Run-in phase to determine the effective OOC dose; responders at week 24 then entered a 36-week randomized controlled treatment (RCT) phase receiving OOC or iSRLs. Acromegaly symptom number and severity (mild to severe, 1-3) were collected. Total score was calculated by summating all severity scores (Acromegaly Index of Severity [AIS]). Symptom results were assessed using total AIS score and proportion of patients experiencing individual symptoms. Results: At beginning of Run-in, average AIS score of 92 randomized patients was 4.52, representative of symptoms experienced while previously receiving iSRLs. After 26 weeks’ OOC treatment at end of Run-in, average AIS score was significantly reduced to 3.46 (P<0.001). More than 80% of patients on OOC improved or maintained AIS score during Run-in compared to baseline. Over this 26-week period, there was a significant reduction in extremity swelling (P=0.01) and fatigue (P=0.03). During the RCT, of patients randomized to OOC (n=55), 73% maintained or improved AIS score, and 75% maintained or reduced overall number of active symptoms. In comparison, 68% of those randomized to iSRLs (n=37) maintained or improved AIS score, and 70% maintained or reduced overall number of active symptoms. Conclusion: Results from MPOWERED show that patients receiving OOC had significant improvement in number and severity of acromegaly symptoms after switching from iSRLs. These findings validate previous results from a phase 3 study of OOC in acromegaly in which patients switching to OOC from iSRLs showed significant reduction in joint pain, extremity swelling, and fatigue.1 1Melmed S, et al. JCEM. 2015;100(4):1699-1708.

Machine Learning-based Prediction Model for Treatment of Acromegaly With First-generation Somatostatin Receptor Ligands

2021 ◽  
Author(s):  
Luiz Eduardo Wildemberg ◽  
Aline Helen da Silva Camacho ◽  
Renan Lyra Miranda ◽  
Paula C L Elias ◽  
Nina R de Castro Musolino ◽  
...  
Keyword(s):  
Machine Learning ◽  
Support Vector Machine ◽  
Prediction Model ◽  
Predictive Value ◽  
First Generation ◽  
Somatostatin Receptor ◽  
Support Vector ◽  
Receptor Ligands ◽  
Somatostatin Receptor Ligands ◽  
Igf I

Abstract Context Artificial intelligence (AI), in particular machine learning (ML), may be used to deeply analyze biomarkers of response to first-generation somatostatin receptor ligands (fg-SRLs) in the treatment of acromegaly. Objective To develop a prediction model of therapeutic response of acromegaly to fg-SRL. Methods Patients with acromegaly not cured by primary surgical treatment and who had adjuvant therapy with fg-SRL for at least 6 months after surgery were included. Patients were considered controlled if they presented growth hormone (GH) <1.0 ng/mL and normal age-adjusted insulin-like growth factor (IGF)-I levels. Six AI models were evaluated: logistic regression, k-nearest neighbor classifier, support vector machine, gradient-boosted classifier, random forest, and multilayer perceptron. The features included in the analysis were age at diagnosis, sex, GH, and IGF-I levels at diagnosis and at pretreatment, somatostatin receptor subtype 2 and 5 (SST2 and SST5) protein expression and cytokeratin granulation pattern (GP). Results A total of 153 patients were analyzed. Controlled patients were older (P = .002), had lower GH at diagnosis (P = .01), had lower pretreatment GH and IGF-I (P < .001), and more frequently harbored tumors that were densely granulated (P = .014) or highly expressed SST2 (P < .001). The model that performed best was the support vector machine with the features SST2, SST5, GP, sex, age, and pretreatment GH and IGF-I levels. It had an accuracy of 86.3%, positive predictive value of 83.3% and negative predictive value of 87.5%. Conclusion We developed a ML-based prediction model with high accuracy that has the potential to improve medical management of acromegaly, optimize biochemical control, decrease long-term morbidities and mortality, and reduce health services costs.

Sign in / Sign up

Export Citation Format

Share Document