July 2024

This prospective study evaluated the liver pre-metastatic niche from liver biopsies taken at the time of pancreatectomy in patients with localized pancreatic ductal adenocarcinoma (PDAC). The liver biopsies were analyzed using metabolomic, transcriptomic, and multiplex imaging approaches. The authors found that liver biopsies from patients with liver metastasis had less T cell lobular infiltration, less steatosis, and higher levels of citrullinated H3, while livers of patients who did not develop liver metastasis were similar to control liver biopsies taken from patients without PDAC. The authors were able to use the liver profiles and a machine-learning-based model to predict the metastatic outcome at the time of surgery with 78% accuracy.

Summary:
PDAC is a deadly disease with a dismal 5-year overall survival rate of 12% that has not significantly changed in 60 years. The only potential cure for PDAC is surgical resection, but this is only available to roughly 25% of cases due to the aggressive and metastatic nature of PDAC. Despite surgical resection, liver metastasis will develop in 40% of patients within the first 3 years after surgery and is almost uniformly fatal. The inability to predict the risk of subsequent metastases and effectively treat it represents a major challenge in the management of PDAC.

In this prospective study, the investigators enrolled 49 patients with resectable PDAC who had not received neoadjuvant therapy and 19 non-PDAC controls (IPMN, NET, and benign) to have a liver biopsy at the time of pancreatectomy. They then followed the 49 patients with PDAC and classified them into early (<6 months), late (>6 months) liver metastasis, extrahepatic metastasis, and disease-free survivors. First, the authors compared RNA transcriptomic profiles from liver biopsies of patients with PDAC to controls. They found that gene signatures associated with inflammation, such as IFNa response, allograft rejection, and monocyte chemotaxis, were higher in patients with PDAC compared to controls. This was supported with immunohistochemistry staining and single-cell RNA sequencing, which showed increased portal/lobular inflammation, T cell abundance, and NETs.

Next, the authors investigated differences between patients who developed liver metastasis versus those that did not (disease-free survivors and extrahepatic metastasis). At the transcriptomic level, patients who developed liver metastasis within 6 months of resection had a high expression of SORT1, which regulates cytokine secretion in myeloid and T cells. SORT1 was associated with a three-fold increased risk of early liver metastasis (HR 2.69, P=0.029). Liver metastasis patients were also less likely to have T cell lobular infiltration and higher levels of neutrophil extracellular traps. Both factors also correlated with time to liver metastasis. At a metabolic level, 15 metabolites distinguished most of the patients who subsequently developed liver metastasis. Of these, low hepatic creatine levels (P=0.04) and high citrullinated H3 (P=0.009) were associated with shorter time to liver metastasis. The authors concluded that patients who develop liver metastasis have a specific immunometabolic profile at the time of surgery.

Finally, the authors used all of the above data to generate a predictive model for metastatic outcome. They generated four models that would predict each of the four metastatic outcomes in a binary fashion. The AUC of the four models ranged from 0.83 to 0.89. The model predicting early recurrence did the best with a 90% sensitivity and 87% specificity. When the models were combined, the overall accuracy of predicting which patient will have which type of recurrence was 78%. The authors ultimately concluded that immunometabolic characterization of peri-operative liver biopsies could predict liver metastasis.

This study's limitations include a small sample size and potential biases from patient selection and pre-operative treatments. Additionally, the predictive models would benefit from further validation in larger cohorts. Lastly, technical variability could affect the reproducibility and reliability of the results.

Bottom Line:
Currently, there is no reliable method to predict metastasis in PDAC, but the authors of this study showed that immunometabolic profiling of peri-operative liver biopsies demonstrated a distinct pre-metastatic niche that could be used to predict liver metastasis.

This prospective, multicenter, double-blinded randomized trial compared the effect of mechanical and oral antibiotic bowel prep (MOABP) to mechanical bowel prep plus placebo (MBP) on patient morbidity following rectal resection. The authors found that compared to MBP, patients in the MOABP group experienced significantly fewer postoperative complications, as measured by the Comprehensive Complication Index, surgical site infections (SSI), and anastomotic leaks.

Summary
For decades, preoperative bowel prep, including combinations of mechanical and/or antibiotics, has been shown to decrease postoperative complications in colorectal surgery. As such, the ASCRS clinical practice guidelines offer a level 1B recommendation for both mechanical and antibiotic prep prior to elective colorectal resection. Surgical site infections (SSI) can affect up to one-third of patients undergoing colorectal surgery and lead to significant patient morbidity, with the risk even more pronounced for those undergoing low rectal or coloanal anastomoses. However, the literature to date contains a paucity of comparative outcomes data on different bowel prep modalities for patients undergoing rectal-only resection. Therefore, the authors performed the “Mechanical Bowel Preparation and Oral Antibiotic vs. Mechanical Bowel Prep Only Prior to Rectal Surgery” (MOBILE2) trial to determine differences in complication rates using either MOABP or MBP alone.

The Finnish study enrolled 610 patients (565 patients in the modified ITT cohort) for randomization between March 2020 and October 2022 across three large university hospitals. Patients were eligible if they had a rectal tumor within 15 cm from the anal verge, were to undergo an elective low anterior resection with colorectal or coloanal anastomosis and were 18 years old or greater. Exclusion criteria included pre-op bowel obstruction, prior ostomy, and intolerance of MBP or antibiotics. The MBP protocol consisted of 2L polyethylene glycol and 1L clear fluid started 2 days before surgery, and the antibiotic regimen included 1g neomycin and 1g metronidazole at 3 pm and 11 pm the day before surgery.

After 1:1 randomization, there were 277 patients in the treatment and 288 in the control group. The primary outcome of the 30-day Comprehensive Complication Index showed significantly fewer complications in treatment versus control (0 vs. 8.66, Wilcoxon effect 0.146; p<0.001) and a lower Clavien-Dindo III or higher complication rate (7.2% vs. 13.9%; p=0.01). Secondary outcomes showed a similar trend for SSI (8.3% vs. 16.7%; OR 0.45 [95% CI 0.27-0.77]) and anastomotic leak (5.8% vs. 13.5%; OR 0.39 [95% CI 0.21-0.72]). There was no difference in length of stay or need for adjuvant therapy. Subgroup analysis showed a reduction in complications in low rectal tumors, patients with no or short-course RT, with protective stoma, and open surgery. There were no differences demonstrated for high rectal tumors or long-course RT. There was only 1 case of Clostridioides difficile, which occurred in the MOABP group.

Limitations to this study include the relatively low rate of minimally invasive surgery performed, which limits generalizability and may inflate the effect size; small sample size in the subgroup analysis may lead to an increased type II error and recall bias may have affected surgeons’ reporting of their SSI rates.

Bottom Line:
The MOBILE2 trial is the first large, prospective, randomized, double-blind placebo-controlled trial to assess the effect of combination mechanical prep with antibiotics on postoperative complications in rectal resections. Compared to MBP, the use of MOABP reduced major complications by almost 50% and decreased SSI and leak rates; mechanical and antibiotic bowel prep should be considered for all eligible patients undergoing rectal cancer resection.

The phase 3 NADINA trial compared neoadjuvant and adjuvant immunotherapy in patients with resectable stage III melanoma. The neoadjuvant group received two cycles of ipilimumab plus nivolumab, followed by a therapeutic lymph node dissection and response-driven adjuvant treatment. This regimen resulted in a 12-month event-free survival rate of 83.7%, significantly higher than 57.2% in the adjuvant group, which received 12 cycles of nivolumab after surgery.

Summary
Recent phase 1-2 trials of patients with clinical stage III melanoma have shown that neoadjuvant immunotherapy may lead to improved event-free survival compared to standard-of-care adjuvant systemic therapy following surgical resection. These include the SWOG1801 trial which demonstrated a 23% improvement in 2-year event-free survival favoring a neoadjuvant/adjuvant approach with pembrolizumab (a PD-1 inhibitor) compared to adjuvant therapy alone. In addition, phase 2 trial data suggests that combination immunotherapy consisting of ipilimumab (a CTLA-4 inhibitor) plus nivolumab (a PD-1 inhibitor) in the neoadjuvant setting is safe and may be even more effective in treating this disease. One prevailing theory regarding the success of neoadjuvant therapy is a stronger and more diverse T-cell response due to the presence of the entire tumor during immunotherapy initiation.

In the current NADINA phase 3, international, randomized controlled trial, investigators compared event-free survival for patients with resectable stage III melanoma undergoing one of two treatment regimens. Patients were randomly assigned to receive either neoadjuvant or adjuvant immunotherapy. The neoadjuvant group received two cycles of ipilimumab and nivolumab, followed by therapeutic lymph node dissection and further treatment based on the pathological response. The adjuvant group underwent surgery first, followed by 12 cycles of nivolumab. The primary endpoint was event-free survival (events including progression to unresectable melanoma before surgery, disease recurrence, and death due to melanoma or treatment).

A total of 423 patients were randomized. There were 28 events in the neoadjuvant group with a median follow-up of 10.6 months and 72 events in the adjuvant group with a median follow-up of 9.9 months. The 12-month event-free survival was 84% in the neoadjuvant group compared to 57% in the adjuvant group, with an adjusted hazard ratio of 0.32 (p<0.001). In the neoadjuvant group, 59% of patients had either a complete or near-complete pathologic response. Among neoadjuvant patients, the 12-month recurrence-free survival was 95% for those with a major pathological response, 76% for those with a pathological partial response, and 57% for those with a pathological non-response. High-grade adverse events related to systemic treatment (such as hypothyroidism and adrenal insufficiency) occurred in 30% of neoadjuvant patients and 15% of adjuvant patients.

The results of the NADINA trial align with previous phase 2 studies, showing durable outcomes and high overall survival rates associated with neoadjuvant immunotherapy in resectable stage III melanoma. This study also supports earlier work which shows that the addition of a CTLA-4 inhibitor to neoadjuvant PD-1 blockade further improves efficacy. This trial also emphasized the importance of evaluating pathological response in guiding treatment. For patients with a major pathologic response to neoadjuvant immunotherapy, adjuvant treatment can be safely omitted, improving quality of life and reducing healthcare costs and potentially offsetting some increased toxicity attributable to neoadjuvant systemic therapy.

Bottom Line
In patients with resectable stage III melanoma, neoadjuvant therapy with ipilimumab and nivolumab followed by surgery significantly improved event-free survival, showing that this is a safe and effective treatment regimen.