eFigure 1. Forest Plot for Overall Survival
eFigure 2. Forest Plot for Cancer-Specific Survival
eFigure 3. Forest Plot for Disease-Free Survival
eFigure 4. Funnel Plot for Overall Survival Analysis
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Petrelli F, Cortellini A, Indini A, et al. Association of Obesity With Survival Outcomes in Patients With Cancer: A Systematic Review and Meta-analysis. JAMA Netw Open. 2021;4(3):e213520. doi:10.1001/jamanetworkopen.2021.3520
Is obesity associated with better prognosis in patients with cancer?
This meta-analysis of 203 studies with more than 6.3 million participants found that obesity was associated with increased overall and cancer-specific mortality, especially among patients with breast, colon, and uterine cancer. In contrast, patients with obesity and renal cell carcinoma, lung cancer, or melanoma had better survival than patients without obesity.
These findings suggest that survival outcomes are poor among patients with obesity and cancer, except in lung cancer and melanoma.
Obesity, defined as a body mass index (BMI) greater than 30, is associated with a significant increase in the risk of many cancers and in overall mortality. However, various studies have suggested that patients with cancer and no obesity (ie, BMI 20-25) have worse outcomes than patients with obesity.
To assess the association between obesity and outcomes after a diagnosis of cancer.
PubMed, the Cochrane Library, and EMBASE were searched from inception to January 2020.
Studies reporting prognosis of patients with obesity using standard BMI categories and cancer were included. Studies that used nonstandard BMI categories, that were limited to children, or that were limited to patients with hematological malignant neoplasms were excluded. Screening was performed independently by multiple reviewers. Among 1892 retrieved studies, 203 (17%) met inclusion criteria for initial evaluation.
Data Extraction and Synthesis
The Meta-analysis of Observational Studies in Epidemiology (MOOSE) and Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guidelines were reporting guideline was followed. Data were extracted by multiple independent reviewers. Risk of death, cancer-specific mortality, and recurrence were pooled to provide an adjusted hazard ratio (HR) with a 95% CI . A random-effects model was used for the retrospective nature of studies.
Main Outcomes and Measures
The primary outcome of the study was overall survival (OS) in patients with cancer, with and without obesity. Secondary end points were cancer-specific survival (CSS) and progression-free survival (PFS) or disease-free survival (DFS). The risk of events was reported as HRs with 95% CIs, with an HR greater than 1 associated with a worse outcome among patients with obesity vs those without.
A total of 203 studies with 6 320 365 participants evaluated the association of OS, CSS, and/or PFS or DFS with obesity in patients with cancer. Overall, obesity was associated with a reduced OS (HR, 1.14; 95% CI, 1.09-1.19; P < .001) and CSS (HR, 1.17; 95% CI, 1.12-1.23; P < .001). Patients were also at increased risk of recurrence (HR, 1.13; 95% CI, 1.07-1.19; P < .001). Conversely, patients with obesity and lung cancer, renal cell carcinoma, or melanoma had better survival outcomes compared with patients without obesity and the same cancer (lung: HR, 0.86; 95% CI, 0.76-0.98; P = .02; renal cell: HR, 0.74; 95% CI, 0.53-0.89; P = .02; melanoma: HR, 0.74; 95% CI, 0.57-0.96; P < .001).
Conclusions and Relevance
In this study, obesity was associated with greater mortality overall in patients with cancer. However, patients with obesity and lung cancer, renal cell carcinoma, and melanoma had a lower risk of death than patients with the same cancers without obesity. Weight-reducing strategies may represent effective measures for reducing mortality in these patients.
Obesity, defined as a body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) greater than 30, is a chronic disease with increasing prevalence around the world, largely contributing to important health issues in most countries.1 Alongside body fat, which is a general risk factor for serious illness (eg, metabolic syndrome), greater cardiometabolic risk has also been associated with the localization of excess fat in the visceral adipose tissue and ectopic deposits.2 Several large epidemiologic studies have evaluated the association between obesity and mortality. In particular, a meta-analysis of 230 cohort studies including more than 30 million individuals3 found that both obesity and overweight were associated with an increased risk of all-cause mortality. Despite the evidence that excess mortality increases with increasing BMI, some studies have reached the conclusion that elevated BMI may improve survival in patients with cardiovascular disease, a phenomenon called the obesity paradox.4
Increased BMI is also associated with an increased risk of multiple cancer types.5 In addition, obesity and overweight may increase cancer mortality.6 During last decades, we have observed a more rapid increase in obesity among adult cancer survivors compared with the general population.7 The mechanisms contributing to higher cancer incidence and mortality may include alterations in sex hormone metabolism, insulin and insulin-like growth factor levels, and adipokine pathways.8,9
Various studies have suggested that patients with cancer and a normal BMI (ie, 20-25) have worse outcomes than patients with obesity. This phenomenon (ie, the obesity paradox) in cancer is not well understood and presents controversial explanations.10-12 Three different meta-analyses have led to different results, in particular in lung and renal cell carcinomas.13-15 In lung cancer, obesity is favorably associated with long-term survival of surgical patients. Moreover, in renal cell carcinoma, an inconsistent association of BMI with cancer-specific survival (CSS) was found. Conversely, breast, ovarian, and colorectal cancer are invariably associated with increased mortality in patients with obesity.16-18 The main explanations for these observations include the general poor health status of patients with very low BMI. Additionally, weight loss may be associated with frailty and other risk factors (eg, smoking).11 In cancer, obesity is also associated with increased efficacy of programmed cell death 1 and programmed cell death ligand 1 (PD-1/PD-L1) blockade in both tumor-bearing mice and patients.12 This updated systematic review and meta-analysis was conducted to evaluate the prognosis of patients with cancer who have obesity vs those without obesity.
We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline and the Meta-analysis of Observational Studies in Epidemiology (MOOSE) reporting guideline19,20 A systematic search was conducted of EMBASE, PubMed, and the Cochrane Library for articles published from database inception until September 30, 2020. The following search terms were used: ((carcinoma or cancer or sarcoma or melanoma or (“Neoplasms”[MESH])) AND (obese OR obesity OR 30 kg/m2 OR “body mass index”) AND (hazard ratio) AND survival AND (multivariate OR cox or multivariable). The reference lists of identified articles were then manually searched to identify potentially relevant omitted citations. Articles that were not published in English were not included.
Retrospective and observational studies (ie, cohort, case-control) or prospective trials were selected when they reported the association of obesity, defined as a BMI of at least 30, with the risk of death (ie, overall survival [OS]), CSS, disease-free survival (DFS), or progression-free survival (PFS) in patients with cancer compared with counterparts without obesity (ie, BMI <30). We placed no restrictions on study setting, size, race, or country. Included studies were limited to those reporting hazard ratios (HRs) and their corresponding 95% CIs. Studies were restricted to adult patients with solid tumors. Hematologic malignant neoplasms were excluded. Short-term survival studies (eg, postsurgical mortality) were also excluded. Baseline-only BMI evaluation was considered (eg, BMI captured at cancer diagnosis in early-stage cancers or at metastatic disease in advanced-stage cancers).
The most up-to-date versions of full-text publications were included. Study selection was performed in 2 stages. First, titles and abstracts were screened; then, selected full-text articles were included according to the eligibility criteria. If pooled analyses of more than 1 study were evaluated for inclusion, the included articles were manually evaluated for duplicate inclusion compared with the other eligible articles. Screening was performed independently by 10 authors (M.G., G.T., A.G., A. Indini, A.C., O.N., V.R., A. Iaculli, L.D., M.S.), and conflicts were handled by consensus with a senior author (F.P.).
Data were collected independently by using a predesigned spreadsheet (Excel version 2007 [Microsoft Corp]). Collected data items included authors, year of publication, study setting and design, median follow-up, treatments received, outcomes, and type of analysis. The primary outcome was OS; secondary end points were CSS and PFS or DFS. Along with data extraction, 1 author (F.P.) assessed study quality according to a modified Newcastle Ottawa Scale (NOS; range 1-9, with 1-3 indicating low quality, 4-6 indicating moderate quality, and 7-9 indicating high quality).21
First, pooled HRs with 95% CIs were estimated using random-effects meta-analysis with the generic inverse-variance method for studies that provided fully adjusted HRs. Inconsistency across studies was measured with the I2 method. Cutoff values of 25%, 50%, and 75% indicated low, moderate, and high heterogeneity, respectively. When I2 was larger than 50%, a random-effects model was primarily used because of the retrospective nature of included studies. To examine heterogeneity, we performed exploratory analyses of predefined subgroups based on type of disease, type of study, duration of follow-up, and race. Additionally, to address potential bias and verify our results, we performed sensitivity analyses using a leave-one-out method and the trim-and-fill method.22 These methods explore whether there are potential dominant studies that may have driven the results. Finally, to investigate the risk of publication bias, we applied the Egger test and visually inspected the funnel plots (ie, the Begg test).23 If the distribution of studies is symmetrical, the meta-analysis most likely does not have problems with publication bias. All statistical tests were 2-sided using a significance level of P < .05. All analyses were carried out using Comprehensive Meta-Analysis software version 3.3.070.
Our literature search yielded 1892 articles, of which 203 (17%) met the inclusion criteria for our overall systematic review of the association of obesity with cancer outcomes (Figure). Most excluded studies did not use the prespecified cutoff value for obesity (ie, BMI values different from 30 in 437 studies) or used a continuous cutoff for risk of death (eg, 1 unit-increase in BMI in 235 studies). Of the 203 articles, 170 (84%) were eligible for inclusion in the systematic review of the association of obesity with OS, 109 (54%) for association with CSS, and 79 (39%) for association with DFS or PFS. Descriptive data for studies included in our meta-analysis are listed in Table 1.12,24-225 Overall, the included studies included a total of 6 320 365 patients. Sample sizes ranged from 41 to 1 096 492 patients, with a median of 1543. Most studies were retrospective in nature (132 studies [63%]); the minority were prospective cohort or observational studies (63 studies [31%]) or pooled analyses or randomized trials (8 studies [4%]). Multivariable analysis was performed in 197 studies. Overall, 136 studies (63%) reported a significant association of obesity with the outcome in at least 1 end point. The mean NOS score was 7 (median, 7.5; range, 5-9), indicating that the overall quality of articles was good.
A total of 170 studies reported data on OS. Because the heterogeneity test showed a high level of heterogeneity (I2 = 79.7%; P < .001) among studies, a random-effects model was used for the analysis. OS among patients with obesity was significantly worse than that among patients without obesity (HR, 1.14; 95% CI, 1.09-1.19; P < .001) (eFigure 1 in the Supplement). The association of obesity with outcomes was independent by other main cancer prognostic factors, including stage (100%), sex (85%), age (100%), race (80%), smoking status (83%), and other comorbidities according to multivariable analysis.
Similarly, obesity was associated with reduced CSS in 109 studies (HR, 1.17; 95% CI, 1.12-1.23; P < .001) (eFigure 2 in the Supplement). Heterogeneity was high (I2 = 73.9%; P < .001), so a random-effects model was used.
In 79 studies, obesity was associated with worse DFS or PFS compared with not having obesity (HR, 1.13; 95% CI, 1.07-1.19; P < .001) (eFigure 3 in the Supplement). Heterogeneity was high (I2 = 73.7%; P < .001), so a random-effects model was used.
A subgroup analysis for OS was performed according to type of disease (Table 2, Table 3, and Table 4). Patients with breast, colorectal, or uterine cancers and obesity had higher overall mortality than those without obesity (breast: HR, 1.26; 95% CI, 1.2-1.33; P < .001; colorectal: HR, 1.22; 95% CI, 1.14-1.31; P < .001; HR, 1.20; 95% CI, 1.04-1.38; P = .01). Patients with obesity and lung cancer, renal cell carcinoma, or melanoma had better survival outcomes compared with patients without obesity and the same cancer (lung: HR, 0.86; 95% CI, 0.76-0.98; P = .02; renal cell: HR, 0.74; 95% CI, 0.53-0.89; P = .02; melanoma: HR, 0.74; 95% CI, 0.57-0.96; P < .001). CSS was decreased in patients with obesity and breast, colorectal, prostate, and pancreatic cancers (breast: 1.23; 95% CI, 1.15-1.32; P < .001; colorectal: HR, 1.24; 95% CI, 1.16-1.32; P < .001; prostate: HR, 1.26; 95% CI, 1.08-1.47; P = .01; pancreatic: HR, 1.28; 95% CI, 1.05-1.57; P = .01). DFS was decreased in patients with obesity and breast, colorectal, prostate, and gastroesophageal cancers (breast: HR, 1.14; 95% CI, 1.1-1.19; P < .001; colorectal: HR, 1.15; 95% CI, 1.01-1.3; P = .01; prostate: HR, 1.29; 95% CI, 1.07-1.56; P < .001; gastroesophageal: HR, 1.62; 95% CI, 1.13-2.32; P < .001). Additional subgroup analyses included type of study (retrospective: HR, 1.07; 95% CI, 1.07-1.18; P < .001; prospective: HR, 1.14; 95% CI, 1.05-1.23; P < .001), duration of follow up (>10 years: HR, 1.16; 95% CI, 0.86-1.58; P = .08; <10 years: HR, 1.23; 95% CI, 0.84-1.63; P = .09), race (non-Asian race: HR, 1.22; 95% CI, 0.86-1.66, P = .06; Asian race: HR, 1.22; 95% CI, 0.74-1.72; P = .09), and stage of disease (early: HR, 1.20; 95% CI, 0.99-1.25; P = .07; advanced: HR, 1.2; 95% CI, 1.12-1.28; P = .01). Regression analysis according to NOS score was not significant.
A funnel plot was used to assess publication bias in the studies evaluating OS in patients with and without obesity. No publication bias was detected by funnel plot inspection (Begg test). Egger test was instead significant (eFigure 4 in the Supplement). According to the trim-and-fill method, 18 studies were placed to the left of the mean, and according the random-effect model, the final result for OS was similar (HR, 1.08; 95% CI, 1.03-1.13). After the leave-one-out procedure, HRs for OS ranged from 1.14 to 1.15.
This meta-analysis found that overall mortality was increased in patients with obesity and breast, colorectal, or uterine cancers. Cancer mortality was increased in breast, colorectal, prostate, and pancreatic cancers. Finally, the relapse rate was increased in breast, colorectal, prostate and gastroesophageal cancers. The obesity paradox, which describes improved cancer and all-cause mortality rates among patients with obesity, was observed in lung cancer and in melanoma; however, these data derive from only 12 studies. We used a categorical BMI definition of obesity (ie, BMI ≥30), because a more standardized definition would permit the comparison and synthesis of studies better than other categories (eg, continuous measures or unit of BMI increase).
The magnitude of effect size was similar for both OS and CSS in breast, colorectal, and lung cancer. This means that obesity may affect both the natural history of cancer and noncancer-related deaths.
Various factors are potentially associated with increased cancer mortality in some malignant neoplasms. Hormonal factors, reduced physical activity, more lethal or aggressive disease behavior, metabolic syndromes, and potential undertreatment in patients with obesity are possible reasons. It is well known that postmenopausal women with higher BMI have an increased risk of breast cancer because of higher estrogen levels resulting from the peripheral conversion of estrogen precursors (from adipose tissue) to estrogen.226 In these patients, weight loss and exercise may reduce cancer risk by lowering exposure to breast cancer biomarkers.227 In colorectal cancer, prediagnosis BMI was associated with increased all-cause, cardiovascular, and colorectal cancer–specific mortality.228 The reason for this association is not presently understood, although insulin, insulin-like growth factors, their binding proteins, chronic inflammation, oxidative stress, and impaired immune surveillance have been supposed to be causative factors.229 In pancreatic cancer, higher prediagnostic BMI was associated with more advanced stage at diagnosis, with 72.5% of patients with obesity presenting with metastatic disease vs 59.4% of patients with reference-range BMI (P = .02) in 2 large prospective cohort studies.161 Lastly, in prostate cancer, obesity may be a consequence of androgen deprivation therapy but seems also associated with more aggressive disease (ie, Gleason score ≥7)230 or more advanced disease at diagnosis.231
Our results showed that patients with obesity and lung cancer had significantly prolonged CSS and OS compared with patients without obesity. When considering these findings, we must take into account that 9 of 11 evaluated studies included patients with advanced and/or metastatic disease. Cancer cachexia mechanisms are not completely defined, but research has shown that the systemic inflammatory status induced either by the tumor or host response is a key moment in the development of cachexia.232 Lung cancers are indeed known to be aggressive, and patients with advanced disease usually have poorer performance statuses and experience significant weight loss at the time of diagnosis, which underlies a systemic inflammatory response.233 In our studies, obesity was positively associated with OS, independent of smoking status, in patients with lung cancer. Interestingly, a post hoc pooled analysis of randomized prospective trials comparing a PD-L1 checkpoint inhibitor (atezolizumab) with docetaxel in patients with advanced non–small cell lung cancer (NSCLC), revealed that the OS benefit for patients with obesity vs those with reference-range BMI was restricted to patients who received immunotherapy; no association was found in the group receiving docetaxel.147 Another study also explored the role of baseline BMI and BMI variation during treatment in a cohort of patients with advanced NSCLC and PD-L1 expression of at least 50% who received first-line pembrolizumab (a PD-1 checkpoint inhibitor) and in a control cohort of patients with NSCLC receiving first-line standard chemotherapy, confirming that the survival benefit for patients with obesity was restricted to those receiving immunotherapy.234
Similar findings have been described in patients with melanoma receiving immunotherapy, and a survival benefit for patients with obesity was reported in the single study205 included in our meta-analysis. However, despite some evidence showing that patients with obesity and melanoma who were receiving immune-checkpoint inhibitors achieved better outcomes,235,236 the association is currently questioned, given that opposite results have been reported in a multicenter study.237
Interestingly, patients with obesity and renal cell carcinoma also had a significantly longer OS compared with the patients without obesity. It has been hypothesized that the perinephric white adipose tissue acts as a reservoir of activated immune cells, with increased characteristics of hypoxia, infiltration of T helper type 1 cells, regulatory T cells, dendritic cells, and type 1 macrophages. However, only 1 of 6 studies included patients who were receiving immunotherapy.238,239
Intriguingly, we found that the association between obesity and better clinical outcomes was confirmed for those malignant neoplasms in which immune checkpoint inhibitors have first (and strongly) proved to be effective; however, studies involving patients receiving immune checkpoint inhibitors are poorly represented in this meta-analysis. Such results might be an epiphenomenon; however, we speculate that white adipose tissue could be considered an immune organ, which somehow plays a role in the antitumor immune response. It has been observed that the adipocyte-derived hormone leptin could alter T cell function, resulting in improved response to anti–PD-1 therapy.12 Moreover, another preclinical study reported that white adipose tissue acts as a reservoir for a peculiar population of memory T cells, which elicit some effective responses in the case of antigenic re-exposure during infections (and why not in case of exposure to cancer-specific antigens?).240 Finally, considering that immune checkpoint inhibitors exert their action within the tumor microenvironment, modulating the interactions between the tumor and the host, it has been proposed that systemic metabolic conditions, including high blood cholesterol, obesity, hyperglycemia and diabetes, atherosclerosis, and hypertension, may represent the epiphenomena of an inflamed patient. Such a patient might be characterized by an enrichment of cytokines and pro-inflammatory mediators (both in the innate and adaptive compartments) and by a condition of T cell exhaustion, with defective cellular-mediated mechanisms. Nevertheless, in these patients, immune checkpoint blockade might be more effective in reversing this immunological anergy both at the tumor and at the systemic levels.241
Patients with obesity are also at increased risk of reduced physical activity. Various studies highlighted this concept. Physical activity decreases over time in patients with obesity.242,243 In particular, physical activity is strictly associated with breast cancer and colorectal cancer mortality.244,245 Therefore physical activity (or inactivity) should be a major target of obesity prevention and treatment in particular for patients with cancer. Type 2 diabetes is strongly associated with obesity in the metabolic syndrome. More than 80% of cases of type 2 diabetes can be attributed to obesity, which may also account for many diabetes-related deaths. The association between BMI and cause-specific mortality was also illustrated in the Prospective Studies Collaboration analysis.246 In the upper BMI range (ie, 25 to 50), each 5-unit increase in BMI was associated with a significant increase in mortality from coronary heart disease, stroke, diabetes, chronic kidney disease, and many cancers. In the same analysis, individuals with BMI less than 22.5 had higher mortality compared with individuals with a BMI of 22.5 to 25. The excess mortality was predominantly associated with smoking-related diseases (ie, respiratory disease and cancer). However, there are no clear recommendations about dosing of chemotherapy in patients with obesity, so caution is recommended for high-risk regimens.247 The hypothesis that a reduced dose according to ideal body weight may lead to a worse outcome cannot be confirmed by prospective studies but may be considered a potential reason for the observed results in some settings (eg, breast cancer). In a pooled analysis of toxic effects in patients with and without obesity, rates of toxic effects were similar or lower in patients with obesity.248
This study has several limitations. First, we combined data for patients with obesity and compared their prognosis with patients with different weights (ie, normal weight or normal weight and overweight). Second, accurate measures of potentially self-reported weight and height are always a challenge in observational studies. The evaluation often takes place before diagnosis, but in some studies the timing of the obesity diagnosis was not described. Patients with obesity have a generally poor prognosis in terms of overall mortality and noncancer mortality, so it seems obvious that their prognosis would be worse than patients without obesity. However, almost all studies provided a multivariate analysis according to main prognostic factor for oncological outcome so that obesity remains generally an independent prognostic factor in patients with cancer. The outcome was almost never adjusted for private medical insurance, but obesity can increase costs for cancer treatment and complications. Therefore, patients with a lower socioeconomic status may have had reduced access to medical facilities (ie, access to anticancer treatments), rehabilitation, or follow-up intensity and therefore had inferior outcomes. Duration of follow-up, treatments received, and countries were heterogeneous even if subgroup analyses did not explain results with these different variables. Furthermore, this meta-analysis compared mortality between patients belonging to a fixed category of obesity (ie, BMI >30), and thus, we are not able to provide an effect size per unit increment.
In this study, the results supported the notion that obesity is a competing risk factor for overall and cancer specific mortality as well as recurrence in various cancers treated with curative intent or for metastatic disease, except for lung cancer and melanoma, in which obesity was associated with reduced mortality (obesity paradox). These results suggest that oncologists should increase their efforts to manage patients in multidisciplinary teams for care and cure of both cancer and obesity. Improving lifestyle factors (eg, physical activity, caloric intake, care and prevention of cardiovascular complications), more intensive follow-ups of cancer in patients with obesity, and adequate dose of medical therapies are all proven measures that may improve prognosis for patients with cancer and obesity.
Accepted for Publication: February 8, 2021.
Published: March 29, 2021. doi:10.1001/jamanetworkopen.2021.3520
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Petrelli F et al. JAMA Network Open.
Corresponding Author: Fausto Petrelli, Oncology Unit, Medical Sciences Department, Azienda Socio Sanitaria Territoriale Bergamo Ovest, Piazzale Ospedale 1, 24047, Treviglio (BG) Italy (firstname.lastname@example.org).
Author Contributions: Dr Petrelli had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Cortellini, Salati, Dottorini, Iaculli, Rampulla, Barni, Cabiddu, A. Ghidini, Zaniboni.
Acquisition, analysis, or interpretation of data: Petrelli, Cortellini, Indini, Tomasello, M. Ghidini, Nigro, Iaculli, Varricchio, Rampulla, Bossi, A. Ghidini.
Drafting of the manuscript: Petrelli, Cortellini, Indini, Tomasello, M. Ghidini, Nigro, Dottorini, Iaculli, Varricchio, Rampulla, A. Ghidini.
Critical revision of the manuscript for important intellectual content: Cortellini, Indini, Salati, Barni, Cabiddu, Bossi, Zaniboni.
Statistical analysis: Petrelli, Cortellini, Dottorini, Varricchio.
Obtained funding: Salati.
Administrative, technical, or material support: Nigro, Rampulla, Bossi, A. Ghidini.
Supervision: Cortellini, Indini, M. Ghidini, Iaculli, Barni, Bossi, A. Ghidini, Zaniboni.
Conflict of Interest Disclosures: Dr Cortellini reported receiving grants from AstraZeneca, Roche, Merck Sharpe and Dohme, Bristol Myers Squibb, Astellas, and Novartis outside the submitted work. Dr Bossi reported receiving grants from Lilly Italia, Novo Nordisk, Bayer, Merck Sharpe and Dohme Italia, Sanofi, and Pikdare outside the submitted work. No other disclosures were reported.
Create a personal account or sign in to: