December 2023

This is a double-blind, randomized, placebo-controlled trial looking at the use of Durvalumab in addition to neoadjuvant chemotherapy in patients with resectable non-small cell lung cancer (NSCLC). They found that adding durvalumab to the treatment regimen for NSCLC prolonged event-free one-year survival, improved pathologic response, and decreased the risk of disease progression, recurrence, and death compared to the placebo group. 

Summary: Five-year recurrence after lung resection for NSCLC is between 30% and 55% and is highly predictive of disease-specific mortality. Chemotherapy used in either the neoadjuvant or adjuvant settings alone for resectable NSCLC has been shown to improve overall five-year survival, but by no more than 5%. Recently, immune checkpoint inhibitors have been approved for use in either neoadjuvant or adjuvant treatment in combination with platinum-based chemotherapy. Durvalumab is a PD-L1 inhibitor established as the international standard for consolidation therapy for patients with unresectable stage III NSCLC. The AEGEAN trial aimed to explore the efficacy of durvalumab used perioperatively (i.e. as neoadjuvant and adjuvant therapy) along with neoadjuvant chemotherapy in patients with resectable NSCLC.

Eligible patients had newly diagnosed, untreated, resectable NSCLC, stages IIa to IIIb (including N2 disease). Documentation of PD-L1 status was required. Patients planned for either sublobar resection or pneumonectomy were excluded. Patients were randomized to receive four cycles of platinum-based chemotherapy and either fixed-dose (1500mg) durvalumab or placebo before surgery. After surgery, patients continued to receive either durvalumab or placebo every 4 weeks up to 12 cycles. Primary endpoints were pathologic complete response and event-free survival, defined as either progressive disease that precluded surgery, progressive disease that was discovered during surgery, local or distant recurrence as assessed by RECIST criteria, or death from any cause. Secondary endpoints were major pathological response, disease-free survival, and overall survival.

A total of 802 patients were randomized in the intention-to-treat population from Asia, Europe, North America, and South America. Median follow up time was 11.7 months. More than 70% of the patients had stage III disease, and half had N2 disease. In total, 77.6% of the patients in the durvalumab group and 76.7% of those in the placebo group underwent surgery. At the 12-month landmark analysis, the percentage of patients with event-free survival was 73.4% in the durvalumab group (95% CI, 67.9 to 78.1) and 64.5% in the placebo group (95% CI, 58.8 to 69.6); at 24 months, event-free survival was 63.3% in the durvalumab group (95% CI, 56.1 to 69.6) and 52.4% in the placebo group (95% CI, 45.4 to 59.0). Event-free survival benefit with durvalumab as compared with placebo was maintained across most subgroups prespecified at baseline. Pathological complete response was seen in a higher proportion of patients in the durvalumab group (17.2%; 95% CI, 13.5 to 21.5) than in the placebo group (4.3%; 95% CI, 2.5 to 6.9). Event-free survival and pathological complete response benefit were observed regardless of stage and PD-L1 expression. Adverse events of maximum grade 3 or 4 occurred in 42.4% of patients with durvalumab and in 43.2% with placebo. The stratified hazard ratio for disease progression, recurrence, or death in the durvalumab group was 0.68 (95% CI: 0.53 – 0.88, p=0.004), meaning this group had a 32% lower risk of these outcomes compared to the placebo group. Limitations of this study include lack of long term follow up data which limits survival analysis as well as the exclusion of patients with certain tumor mutations and translocations (namely EGFR and ALK), which inevitably makes these results less generalizable. 

Bottom Line: In patients with resectable NSCLC, pre- and post-operative durvalumab plus neoadjuvant chemotherapy was associated with significantly greater event-free survival and pathological complete response as compared to neoadjuvant chemotherapy alone.  

This multi-center randomized controlled trial (RCT) sought to determine if tight glycemic control in critically ill patients not on total parenteral nutrition (TPN) improved various outcomes in the intensive care unit (ICU). They found that, even with the use of a computer algorithm to enforce glycemic goals, there was no difference in length of ICU stay or risk of death between patients with tight versus conventional glucose goals.  

Summary: The relationship between blood-glucose level and outcomes has been extensively studied in the ICU population. While many prior RCTs have shown the deleterious effects hyperglycemia as it relates to complications and mortality in the patients with critical illness, subsequent RCTs have been unable to replicate this, with a large trial showing increased mortality attributed to severe hypoglycemia in this patient population. One potential contributing factor to this discordance was due to the use of parenteral nutrition, which is thought to confound results, as well as the lack of standardized glucose measurements in tight glucose control trials. 

Therefore, to address these limitations, the goal of this multi-center RCT was to determine if the use of a computer algorithm called LOGIC-insulin, which guides bedside nurses on how to adjust insulin infusions, to enforce tight glucose control (defined as 80-110 mg/dL) in critically ill patients NOT receiving parenteral nutrition would decrease ICU needs as compared to those with conventional glucose control (defined as 180-215 mg/dL). All the patients received enteral nutrition as soon as possible, but parenteral nutrition was initiated only after 1 week in the ICU. Primary outcome was length of time that ICU care was needed, defined as time to discharge alive from ICU or time until readiness for discharge from ICU. 

Overall, 9230 patients were included in the study, 4622 of which were assigned to the tight glucose control group between September 2018 and August 2022 in 11 ICUs across three hospitals in Belgium. Similar to other RCTs by this group, these patients were overwhelming cardiothoracic surgery patients (about 45% in each arm). Overall, patients with tight glucose control did not have a decreased length of stay in the ICU (HR 1.00, 95% CI: 0.96 – 1.04, p=0.94) but were more slightly more likely to exhibit a hypoglycemia event, although not statistically significant (1.0% vs. 0.7%, RR = 1.52, 95% CI: 0.97 – 2.39). There was also no difference in mortality among the two groups (10.1% vs 10.5%, p=0.51). Secondary outcomes, which included acute kidney injury (AKI), liver dysfunction, and new infection were similar between groups, except a slightly decreased rates of AKI in the tight control group (8.6% vs. 7.2%, RR 0.85, 95% CI 0.73 – 0.97) including need for renal replacement therapy (5.8% vs. 4.7%, RR 0.82, 95% CI: 0.68 – 0.98) and decreased markers of cholestatic liver dysfunction.

There are several limitations of this study, first of which includes the heterogeneity of the patient population that may be skewing results. More importantly, the median glucose level between the two groups were significantly different at 107 vs. 140 mg/dL, but whether that 30 mg/dL difference is large enough to be clinically significant is unclear. Another approach might be to consider the impact of glucose variability during the stay on outcomes, a facet that has previously been associated with poor outcomes. Similarly, it might be useful to evaluate insulin rate or degree of insulin resistance via HOMA-IR and outcomes across the groups. Regardless, this study only reinforces that work should be focused on the potential pathophysiology driving the association between high glucose levels >200 mg/dL and outcomes. The trend toward improved kidney and liver function with lower glucose levels is consistent with the literature and emphasizes the importance in understanding the mechanisms underlying hyperglycemia’s persistent association with bad outcomes - which highlights a great place to focus future efforts. 

Bottom line: Tight glycemic control, as directed by a computer-based algorithm, does not decrease the risk of prolonged ICU stay or death in patients with critical illness who are not receiving parenteral nutrition.

This is a retrospective cohort study evaluating the impact of social vulnerability on surgical evaluation for primary hyperparathyroidism. The authors found that patients with high social vulnerability indices were less likely to be evaluated by a surgeon for primary hyperparathyroidism. 

Summary: An estimated 100,000 people in the United States (US) are diagnosed with primary hyperparathyroidism (PHPT) annually. While surgery is the only effective treatment, up to 80% of patients are never referred for surgical evaluation, and rates of referral are influenced by social determinants of health (SDOH). The CDC’s social vulnerability index (SVI) uses US census data to quantify SDOH based on 15 social factors organized into 4 subthemes: socioeconomic status (SES), household characteristics, minority status and language, housing type and transportation. This study sought to evaluate the impact of SVI on time-to-surgeon evaluation for PHPT and identify which SVI subthemes contribute most to disparities. 

This is a retrospective cohort study of patients in a Massachusetts institutional electronic health record data registry (2010-2018) diagnosed with a first incident hypercalcemia (defined as serum calcium greater than or equal to 10.4 mg/dL) and a subsequent diagnosis of PHPT (defined as parathyroid hormone (PTH) >65 pg/mL). Additionally, patients with normo-hormonal hyperparathyroidism were identified using ICD codes and also included in the cohort. The main exposure was quartile rankings of overall SVI and SVI subthemes, with higher quartile corresponding to greater social vulnerability. The primary outcome was time-to-surgeon evaluation, calculated as date of initial elevated calcium to date of surgeon evaluation in days. Log-rank testing and multivariable cox proportional hazards were used to compare time-to-surgeon evaluation across SVI quartiles.

A total of 1082 patients were included, the majority of which were female (74%) and White race (67%) with a mean age of 62.9 years. Of the included patients, 27.5% were evaluated by a surgeon and 23.1% underwent parathyroidectomy. Patients in higher SVI quartiles were more likely to be Black race and had more comorbidities. Patients with higher SVIs were less likely to be evaluated by a surgeon (Q1 SVI: 31.1% vs Q2 SVI: 31.41% vs Q3 SVI: 25.93% vs Q4 SVI: 21.92%, p = .03). Patients in the highest SVI quartile had a 33% lower estimated rate of surgeon evaluation and were seen an average of 67 days later than patients in the lowest SVI quartile (aHR: 0.67, 95% CI: 0.47-0.97, p = 0.032).  There were specific subthemes [SES, minority status and language, and housing type and transportation] that had a higher impact on the odds of surgical evaluation. Within these subthemes, patients with the highest vulnerability (Q4) had a 35-37% lower estimated rates of surgeon evaluation compared with patients with the lowest vulnerability (Q1). While the single institution nature of this study may limit generalizability, Massachusetts has one of the lowest uninsured rates, suggesting that other states may have even lower and more disparate rates of surgical evaluation for PHPT. 

Bottom Line: Surgical evaluation for PHPT in a Massachusetts cohort was disproportionately lower in patients with higher social vulnerability. Targeted interventions to identify vulnerable patients and address factors contributing to social vulnerability may serve to both mitigate disparities and increase overall rates of surgical evaluation for PHPT.