Neoadjuvant treatment of pancreatic ductal adenocarcinoma: Whom, when and how

Abstract

Pancreatic ductal adenocarcinoma (PDAC), which is notorious for its aggressiveness and poor prognosis, remains an area of great unmet medical need, with a 5-year survival rate of 10% - the lowest of all solid tumours. At diagnosis, only 20% of patients have resectable pancreatic cancer (RPC) or borderline RPC (BRPC) disease, while 80% of patients have unresectable tumours that are locally advanced pancreatic cancer (LAPC) or have distant metastases. Nearly 60% of patients who undergo upfront surgery for RPC are unable to receive adequate adjuvant chemotherapy (CHT) because of postoperative complications and early cancer recurrence. An important paradigm shift to achieve better outcomes has been the sequence of therapy, with neoadjuvant CHT preceding surgery. Three surgical stages have emerged for the preoperative assessment of nonmetastatic pancreatic cancers: RPC, BRPC, and LAPC. The main goal of neoadjuvant treatment (NAT) is to improve postoperative outcomes through enhanced selection of candidates for curative-intent surgery by identifying patients with aggressive or metastatic disease during initial CHT, reducing tumour volume before surgery to improve the rate of margin-negative resection (R0 resection, a microscopic margin-negative resection), reducing the rate of positive lymph node occurrence at surgery, providing early treatment of occult micrometastatic disease, and assessing tumour chemosensitivity and tolerance to treatment as potential surgical criteria. In this editorial, we summarize evidence concerning NAT of PDAC, providing insights into future practice and study design. Future research is needed to establish predictive biomarkers, measures of therapeutic response, and multidisciplinary strategies to improve patient-centered outcomes.

Key Words: Pancreatic adenocarcinoma; Neoadjuvant treatment; Chemotherapy; Upfront surgery; Radiotherapy; Response evaluation

Core Tip: Pancreatic ductal adenocarcinoma, which is notorious for its aggressiveness and poor prognosis, remains an area of great unmet medical need. The most important determinant of survival is surgical resection. Nearly 60% of patients who undergo upfront surgery for resectable tumours are unable to receive adequate adjuvant chemotherapy (CHT). An important paradigm shift to achieve better outcomes has been the sequence of therapy, with neoadjuvant CHT preceding surgery. Three surgical stages have emerged for the preoperative assessment of nonmetastatic pancreatic cancers: Resectable, borderline-resectable and unresectable locally advanced tumours. In the development of new neoadjuvant and induction strategies, the distinct molecular and biological characteristics of the various pancreatic cancer subgroups need to be integrated to optimize the selection and sequencing of both established and novel treatment modalities that may improve survival outcomes.

INTRODUCTION

Pancreatic ductal adenocarcinoma (PDAC), which is notorious for its aggressiveness and poor prognosis, remains an area of great unmet medical need, with a 5-year survival rate of 10% - the lowest of all solid tumours. Surgical resection is the most important predictor of survival (due to lower morbidity and death rates) along with postoperative adjuvant treatment. The sole potentially curative treatment is surgical resection; nevertheless, the 5- and 10-year overall survival (OS) rates for surgery by itself are quite low, at 10.4% and 7.7%, respectively. The actual rates are even lower, as these numbers represent the “best of the best” in clinical trial patients; real-world results are much worse[1].

Merely 20% of patients are diagnosed with resectable pancreatic cancer (RPC) or borderline RPC (BRPC) illness, whereas the remaining 80% have tumors that are unresectable, have distant metastases, or are locally advanced pancreatic cancer (LAPC)[2]. Single-agent adjuvant chemotherapy (CHT) increases the 5-year OS rate for upfront, locally resectable PDAC (about 20% of all patients) from 8% for surgery alone to 16%-21% with adjuvant therapy[3-6]. Gemcitabine plus capecitabine in unselected patients, including those older than 80 years, provided a median OS (mOS) of 28 months. Modified folinic acid, fluorouracil, irinotecan, and oxaliplatin (mFOLFIRINOX) in selected patients younger than 79 years of age increased 3-year OS to 63.4% and mOS to 54.4 months[7,8].

Nearly 60% of patients who undergo upfront surgery for RPC are unable to receive adequate adjuvant CHT due to postoperative complications and early cancer recurrence[9]. Just 7% of Medicare patients, for instance, were able to receive all prescribed adjuvant CHT, according to a recent study. Although it is not common to finish because of the previously described reasons, adjuvant CHT for pancreatic cancer patients who have received upfront surgical resection is associated with a greater chance of survival. Patients who undergo appropriate surgical resection for pancreatic cancer often experience cancer recurrence and are not cured, even with aggressive adjuvant management[10].

STAGING CRITERIA

Radiographically localized pancreatic adenocarcinoma is staged based on anatomical tumour-vessel interfaces into RPC, BRPC, and LAPC. Three fundamental ideas form the basis of these classifications: (1) Increasing vascular involvement may raise the risk of harboring radiographically occult metastatic disease; (2) Margin negative resection is more frequently linked to long-term survival; and (3) Margin positivity increases with increasing degree of vascular involvement[11]. The complexity and variability of classification based on these criteria is demonstrated by the fact that, even in high-volume institutions, approximately 30% of patients were misclassified anatomically in a prospective multicenter clinical trial[12]. Three subsets were developed by the MD Anderson Cancer Center Pancreatic Cancer Group (BR-A, BR-B, and BR-C) to evaluate patients with BR-PDAC and determine resectability using parameters other than anatomic criteria. The anatomic criteria for BR disease are met by BR-A disease, and it lacks any additional characteristics that would prevent surgical resection. The biological characteristics of extra-pancreatic disease, such as biopsy-proven lymph node metastases, elevated carbohydrate antigen 19-9 (CA 19-9) levels (1000 U/mL), and indeterminate liver lesions, are indicative of BR-B disease. Individuals diagnosed with BR-C illness form a conditional group that takes into account comorbidities such as advanced age (> 80 years) or an ECOG performance status of 2 or higher[13].

The systemic inflammatory response of cancer cells is involved in cancer progression and malignant transformation[14]. An elevated neutrophil-to-lymphocyte ratio is associated with poor OS and poor cancer-specific survival in patients with advanced-stage disease, and it is predictive of survival after resection of early-stage PDAC[15]. In PDAC patients receiving neoadjuvant treatment (NAT), the systemic inflammation response index (SIRI) can predict the survival of patients following resection and neoadjuvant chemotherapy (NACT), while the post-treatment systemic immune-inflammation index may be a useful prognostic marker[16]. Changes in the SIRI were associated with OS, but the preoperative SIRI was correlated with disease-free survival[17].

Treatment paradigms must be modified in light of advances in biomarkers, such as the discovery of circulating tumor DNA (ctDNA), based on staging criteria that take into consideration morphological restrictions, the biological predisposition for distant failure, and patient performance status[18]. Furthermore, circulating tumour cells can be isolated in the bloodstream of patients with early-stage pancreatic cancer, and their levels are correlated with the probability of survival[19].

SEQUENCE OF THERAPY

A significant paradigm change that has improved outcomes is the order in which therapies are administered, with NACT coming before surgery. Three surgical stages have emerged for the preoperative assessment of nonmetastatic pancreatic cancers: RPC, BRPC, and LAPC. Identifying patients with aggressive or metastatic disease during initial CHT, reducing tumour volume before surgery to improve the rate of margin-negative resection (R0 resection, a microscopic margin-negative resection), lowering the rate of positive lymph node occurrence at surgery, offering early treatment of occult micrometastatic disease, and evaluating tumour chemosensitivity and tolerance to treatment as potential surgical criteria are all part of the main objective of NAT, which aims to improve postoperative outcomes through an enhanced selection of candidates for curative-intent surgery. Selecting patients with a high chance of undergoing margin-negative resection based on response evaluation represents one of the most difficult scenarios with NAT[20]. The primary risk associated with neoadjuvant therapy is that about 30% of patients will develop metastatic disease while receiving treatment. Thus, a crucial unmet need is instruments to help physicians choose first-line CHT regimens that are more effective[21].

As such, it is critical to carefully weigh the various therapeutic options and customize them in light of clinically meaningful biomarkers. We can improve our capacity to choose patients who will most likely benefit from NACT by finding strong predictive markers. Additionally, the discovery of prognostic markers can help guide treatment choices and offer insightful information about the overall course of the disease.

CURRENT TREATMENT TRENDS IN NAT

Due to a lack of randomized phase III trials, NACT for RPC is not currently included in clinical guidelines. However, it may be an alternative to surgery first for patients with clearly resectable disease who only have suspicious imaging findings of metastasis, CA 19-9 levels suggestive of metastatic disease, or who require medical optimization before surgery. Some studies support the use of NAT for treating resectable tumours[22,23], and others have shown that NAT may be considered for patients with resectable tumours and high-risk features[24,25]. A retrospective study aimed at exploring patient subgroups with high-risk resectable PDAC for selecting candidates who may benefit from NAT reported that NAT is an effective treatment for patients with resectable PDAC, particularly when the tumour is in contact with the portal vein/superior mesenteric vein and has a CA 19-9 concentration ≥ 150 U/mL[26]. Propensity score matching analysis of retrospectively analyzed early-stage PDAC patients showed that in patients with early-stage resected pancreatic head adenocarcinoma, NAT followed by resection demonstrated a significant survival benefit when compared with upfront surgery, while the analysis, which represented an updated nationwide retrospective study, also supports multiagent NACT vs upfront surgery[27,28]. On the other hand, prospective randomized studies like NORPACT1 (neoadj mFOLFIRINOX 4 cycles and adj mFOLFIRINOX 8 cycles vs adj mFOLFIRINOX 12 cycles)[29], PANACHE01-PRODIGE48 (neoadj mFOLFIRINOX 4 cycles and adj chemo 8 cycles vs neoadj FOLFOX 4 cycles and adj chemo 8 cycles vs adj chemo 12 cycles)[30], and SWOG S1505 (neoadj mFOLFIRINOX 12 wk and adj mFOLFIRINOX 12 wk vs neoadj gem/nab-P 12 wk and adj gem/nab-P 12 wk) did not confirm the usefullness of NACT[31]. Additionally, there was no evidence supporting the use of NACT in the PREOPANC-2 study, which used mono-agent neoadjuvant chemoradiotherapy (NACRT)[32]. However, neoadjuvant therapy in these patients serves as a biological time-test and increases the delivery of multimodal therapy. Neoadjuvant therapy was not expected to be beneficial in patients with RPC, according to a meta-analysis of 111 cohort series neoadjuvant response rates and resection rates. However, roughly one-third of patients with nonresectable tumors could be anticipated to have a resectable tumor after neoadjuvant therapy, with a similar survival rate to those with primary resectable tumors[25]. Ongoing and future trials, such as Alliance A021806 (perioperative mFOLFIRINOX vs adjuvant mFOLFIRINOX)[33] and PREOPANC-3 (perioperative mFOLFIRINOX vs adjuvant FOLFIRINOX)[34], are intended to provide additional insight into the function of neoadjuvant therapy in patients with RPC.

A short course of neoadjuvant therapy, usually administered for two months before surgery, boosts the 12-month OS to roughly 77% in patients with borderline-resectable disease (with R0/R1 resection rates of 64%-85%), compared with 40% with upfront surgery (with a resection rate of 75%). An 18-month OS of 67% can result from a prolonged course of NACT, often lasting 4 months. Additionally, after 4-6 months of induction combination CHT with or without radiation therapy, 20% of patients with unresectable nonmetastatic PDAC may undergo resection, even in the absence of a discernible radiological response, improving their OS[35]. An optimal NAT regimen has not been agreed upon due to a lack of data from randomized clinical trials[36]. The two most efficient CHT regimens for PDAC are FOLFIRINOX and gemcitabine-nab-paclitaxel (GNP). For patients with pancreatic adenocarcinomas that were both resectable and borderline resectable, neoadjuvant FOLFIRINOX vs neoadjuvant gemcitabine-based CRT did not show any OS benefit, according to the first study’s phase III PREOPANC-2 trial data. Specifically, the mOS was 21.9% and 21.3% [hazard ratio (HR) = 0.87; 95% confidence interval (CI): 0.68-1.12; P = 0.28]. Between treatment arms, resection rates (77% vs 75%, respectively; P = 0.7) and serious adverse rates (49% vs 43%, respectively; P = 0.26) were likewise comparable[32]. The resection rate, R0 resection rate, and 2-year OS did not differ between the groups, according to the SWOG-1505 study, which assessed mFOLFIRINOX and GNP in the NAT setting of RPC. The toxicity profiles of the GNP and mFOLFIRINOX groups in this investigation did not significantly differ, which is noteworthy[31]. The NAPOLI 3 trial, which included liposomal irinotecan in the NALIRIFOX regimen, showed statistically significant and clinically meaningful improvements in OS and progression-free survival (PFS) compared with nab-paclitaxel and gemcitabine in patients with metastatic PDAC who had not previously received treatment in the metastatic setting. However, FOLFIRINOX is not superior in head-to-head comparisons[37]. According to the phase II nITRO trial, perioperative NALIRIFOX is safe and effective, and it should be further studied in randomized trials that contrast it with current CHT regimens for patients with BRPC and LAPC as well as standard surgery upfront followed by adjuvant therapy in patients with RPC[38].

Prospective randomized phase II and III trials exclusively enrolled patients with LAPC in the LAP-07 trial only, and other data supporting NAT in LAPC patients were derived from retrospective reports[39]. Apart from LAPC, which represents a true unresectable tumour, patients who suffer liver oligometastasis from PDAC may be candidates for surgery after NACT according to a meta-analysis based on small studies without randomized controlled trials and according to the opinions of one-third of the experts[40-42].

The role of radiation or chemoradiation during NAT for BRPC remains unclear. There is considerable debate on the best NAT regimen, and it is still uncertain whether is better for BRPC: NACT or NACRT[35]. CRT and maintenance CHT were examined in 449 patients in the LAP-07 study, who did not have progressing illness after CHT alone. After receiving either gemcitabine and erlotinib or gemcitabine alone for four months (the first randomization), 269 participants were randomly assigned to receive either CHT or CRT for an additional two months of treatment. Although CRT was linked to delayed locoregional development, the mOS in the CRT group did not improve[39]. When FOLFIRINOX was the primary induction therapy, the CONKO-007 trial yielded remarkably comparable outcomes. Without impacting OS or PFS, the addition of CRT raised the pathological complete response rate[43]. Some theoretical advantages of new radiotherapy techniques over conventional radiation include shorter treatment times, more focused treatment fields, higher biological effective doses, and the potential for better sparing of adjacent organs at risk. One such technique is stereotactic body radiotherapy, which targets primary tumors with minimal margins and high single doses in few fractions. The problem of delivering ablative doses while minimizing severe toxicity may potentially be solved by magnetic resonance-guided radiotherapy, or MRgRT[44].

RESPONSE EVALUATION

One of the greatest difficulties faced when dealing with NAT is identifying patients who are highly likely to undergo margin-negative resection based on response evaluation. It is widely recognized that conventional imaging has its limitations when it comes to restaging tumors[45]. PDAC is characterized by a significant presence of desmoplastic stroma, accounting for a substantial portion of the tumor volume, often estimated to be as high as 90%. This stroma is believed to primarily arise from cancer-associated fibroblasts[46]. The Response Evaluation Criteria in Solid Tumours have certain limitations when it comes to PDAC with a high presence of desmoplastic stroma. This is because commonly used imaging methods like computed tomography or magnetic resonance imaging cannot accurately distinguish between fibrosis caused by NACT and viable tumor tissue that remains. As a result, it becomes difficult to accurately determine the resectability of the tumor based solely on imaging studies conducted after neoadjuvant therapy. Numerous studies have also highlighted that the radiological appearance following neoadjuvant therapy does not necessarily reflect the patient’s actual response to the treatment[47].

Numerous institutions have made efforts to assess the biochemical response using CA 19-9, a widely recognized tumor marker in PDAC. Nevertheless, there has been a lack of consistency in the evaluation criteria employed by different institutions[48]. Moreover, the evaluation of treatment outcomes becomes complex when there are disparities between radiological and biochemical responses[49]. Yun et al[50] categorized patients into three groups based on their response to NACT and prognosis: Biochemical responders (BR+), BR-/radiology-only responders (RR+), and nonresponders (BR-/RR-). The research revealed that the 3-year survival rate differed significantly among the BR+ group (71.0%), the BR-/RR+ group (53.6%), and the BR-/RR- group (33.1%) (P < 0.001). Furthermore, the response to NACT was highlighted as a crucial factor in predicting recurrence risk, with HR of 2.15 (95%CI: 1.19-3.88; P = 0.011) for BR-/RR+ patients and 3.82 (95%CI: 2.41-6.08; P < 0.001) for BR-/RR- patients when compared to BR+ patients. Irrespective of the response to NACT, patients who underwent adjuvant CHT exhibited a significantly improved 3-year OS rate in comparison to those who did not[50]. Newhook et al[51] proposed innovative classification systems A (progressively decreasing to normal), B (fluctuating bidirectionally towards normal), C (consistently within normal range), D (decreasing but not normalizing), and E (increasing without normalization) based on the dynamics of CA 19-9. The study findings indicated that categories A and B, where CA 19-9 levels eventually normalize post NACT, are linked to favorable prognoses, while types D and E, where CA 19-9 levels remain elevated post NACT, are associated with poorer outcomes. Hence, apart from CA 19-9, investigations should be carried out to assess the efficacy of NACT by examining glucose metabolic activity or utilizing more sensitive biomarkers like ctDNA[52-54].

The efficacy of adjuvant CHT in patients who have undergone NACT remains a topic of debate. Drawing comparisons between the outcomes of adjuvant and neoadjuvant trials poses a challenge due to the fact that adjuvant trials typically involve only those patients with resected pancreatic cancer who are deemed suitable for adjuvant therapy without any signs of disease progression. On the other hand, neoadjuvant studies generally encompass all patients with RPC, as determined by diagnostic imaging, prior to the initiation of any treatment. Therefore, comparisons of the survival outcomes of patients in neoadjuvant and adjuvant trials are difficult if not impossible[55,56].

The development of cytotoxic treatments for various phases of pancreatic cancer has predominantly relied on empirical methods. A reductionist strategy aimed at creating therapies that target crucial genetic mutations has seen minimal achievements. Nevertheless, several elements of neoadjuvant therapy for PDAC, such as the most effective treatment plan, the incorporation of radiotherapy, and the identification of suitable candidates for neoadjuvant therapy, still require further clarification. Since other tumors with a much better prognosis have a greater number of average mutations, such as breast and colorectal cancers, and because the genetic spectra of colorectal, brain, and pancreatic tumors are similar, the average number of genetic alterations in PDAC tumors is insufficient to explain the abnormally poor prognosis and response to therapy[57,58].

There is accumulating research suggesting a relationship between pancreatic cancer subtype (classical vs basal-like) and treatment response[59]. Basal-like PDAC was resistant to FOLFIRINOX, paclitaxel, and tyrosine kinase inhibitors, whereas the classical transcriptional subtype was more susceptible to epidermal growth factor receptor suppression by erlotinib. These data significantly underscore the need for molecular subtyping in therapeutic decision-making in patients with PDAC[60].

CONCLUSION

The development of new neoadjuvant and induction strategies must take into account the distinct molecular and biological characteristics of the various pancreatic cancer subgroups, optimizing the selection and sequencing of both established and novel treatment modalities to improve survival rates. We can improve our capacity to choose patients who will benefit most from NACT by identifying strong prognostic markers. Moreover, prognostic markers can provide valuable insight into the overall disease course and illuminate treatment choices. Future studies are wanted to set up predictive biomarkers, measures of healing response, and multidisciplinary techniques to enhance patient-targeted outcomes.