Surgical indication for and desirable outcomes of conversion surgery in patients with initially unresectable pancreatic ductal adenocarcinoma

Abstract Aim of this review is to propose an acceptable surgical indication for conversion surgery in patients with initially unresectable (UR) pancreatic ductal adenocarcinoma (PDAC) by considering desirable outcomes, including resectability, overall survival (OS), and disease‐free survival (DFS). A comprehensive literature search of PubMed was conducted through July 15, 2019. Eligible studies were those reporting on patients with UR‐PDAC who underwent surgery. We excluded case reports with fewer than 10 patients, insufficient descriptions of survival data, and palliative surgery. When patients with UR‐PDAC with no progression after chemo(radiation) therapy were offered surgical exploration, resectability and median survival time (MST) of those who underwent conversion surgery ranged from 20% to 69% (median, 52%) and from 19.5 to 33 months (median, 21.9 months), respectively. When conversion surgery was carried out in patients with expected margin‐negative resection or with clinical response by Response Evaluation Criteria In Solid Tumors (RECIST), resectability and MST ranged from 18% to 27% (median, 20%) and from 21 to 35.3 months (median, 30 months), respectively. Among patients who underwent conversion surgery based on clinical response and decreased CA19‐9 level after multimodal treatment, resectability and MST ranged from 2% to 24% (median, 4.1%) and from 24.1 to 64 months (median, 36 months), respectively. Decreased CA19‐9 level was a predictor of resectability, OS and DFS by multivariate analysis. In conclusion, decision‐making for conversion surgery based on clinical response and decreased CA19‐9 level after multimodal treatment may be appropriate. With regard to desirable outcomes of OS and DFS, conversion surgery may provide improved survival for patients with initial UR‐PDAC.


| INTRODUC TI ON
Pancreatic ductal adenocarcinoma (PDAC) continues to have a dismal prognosis with a 5-year survival rate of <5%, even in the modern era. 1,2 Most (70%-80%) patients with PDAC have unresectable (UR) disease, which is subclassified according to the status of distant metastasis-locally advanced disease (UR-LA) and metastatic disease (UR-M), such as distant organ metastasis and non-regional lymph node metastasis. Recent implementation of new regimens, such as FOLFIRINOX 3 and gemcitabine + nab-paclitaxel, 4 has provided better clinical response rates, ranging from 23% to 31.6%, and median survival time (MST), ranging from 8.5 to 12 months, even in patients with metastatic PDAC. Recently, conversion surgery, an additional surgery during multimodal therapy in patients with initial UR-PDAC, has been introduced with the goal of prolonging short-and long-term survival. Number of publications on conversion surgery has increased in recent years.  Several review articles [31][32][33][34][35][36] have reported high resectability rates, high margin-negative resection rates, and high negative lymph node rates in patients who underwent conversion surgery with acceptable mortality and morbidity. MST in patients with initial UR-PDAC who underwent conversion surgery was better than that of patients who did not undergo conversion surgery. However, most publications have described unclear surgical indications and varying rates of resectability, overall survival (OS), and disease-free survival (DFS).  As evidence-based guidelines for the management of UR-PDAC are lacking, this review aims to propose an optimal surgical indication considering desirable outcomes of conversion surgery, with special consideration to resectability, OS, and early recurrence rate.

| Search strategy and data sources
Identification of eligible studies was carried out through a search of PubMed (MEDLINE) through 15 July 2019. The following search terms were used: "(unresectable pancreatic ductal adenocarcinoma OR unresectable pancreatic cancer) AND (pancreatectomy OR surgical resection)". Finally, the reference lists of eligible studies were assessed manually to detect any potentially relevant articles ("snowball" procedure).

| Inclusion and exclusion criteria
Eligible studies were those reporting on patients with histologically confirmed unresectable PDAC who underwent surgery after multimodal therapy, including chemotherapy/radiation therapy. Exclusion criteria were as follows: (i) irrelevant studies, (ii) editorials and letters to the editor, (iii) non-English articles, (iv) case reports including fewer than 10 patients undergoing surgical resection, (v) insufficient description of survival data, and (vi) studies involving treatment mainly by ablative or non-surgical technologies.

| Data extraction and tabulation
Two authors (S.S. and T.Y.) conducted data extraction. Variables of interest included general study characteristics (eg, study period, study design, number of patients, resectability), regimens of multimodal therapy and percentages of patients who received them, surgical indication, OS and DFS, and predictive factors for surgical outcome. Data were tabulated when possible. Discordant judgment was resolved by discussion and consensus.
Although surgical indication and resectability varied, MST did not seem to vary according to resectability status.
a Patients with clinical response for ≥6 mo after multimodal therapy.

| Resectability
Five articles identified prognostic factors for resectability in patients with initial UR-PDAC. 7

| D ISCUSS I ON
On imaging studies, the majority of PDAC is classified as UR disease.  identifying predictors of OS also showed decreased CA19-9 level as a prognostic factor for OS. 10,16,19,22,[24][25][26]30 All three articles reporting prognostic factors for DFS showed that CA19-9 response was a prognostic factor. 14,16,25 Among them, the Heidelberg group clearly showed that a post-chemotherapy CA19-9 level <100 U/mL was a favorable prognostic factor for OS and a post-chemotherapy CA19-9 level ≧100 U/mL was a predictor of poor DFS in 280 patients with initial UR-PDAC, including BR in 6%, UR-LA in 68%, and UR-M (oligometastasis) in 26%. 25 They also showed that a post-chemotherapy  16,19 In these situations, CA19-9 level <100 U/mL or 150 U/mL in UR-PDAC can be a reliable marker for conversion surgery ( Figure 1). Moreover, use of diffusion-weighted magnetic resonance imaging (MRI) 39

| Desirable outcomes for conversion surgery and future perspectives
Recent chemotherapy regimens, such as FOLFIRINOX and gem + nab-PTX, provide better MST of 24.2 months in UR-LA 37 and 8.5-12 months in UR-M. 3,4 Although conversion surgery is expected to prolong survival, we should definitely recognize that the early recurrence rate (within 6 months) after conversion surgery is approximately 30%. 14,16,25 In this situation, patients cannot expect a longer survival relative to non-surgical patients, and conversion surgery may simply be a surgical injury for patients, because extensive pancreatectomy has a high risk of mortality and morbidity. 36 The early recurrence rate should be decreased as much as possible in patients who undergo conversion surgery. From the prognostic point of view, desirable outcomes of an MST of 36 months in patients with UR-LA and 24 months in patients with UR-M and less than a 20% incidence of early recurrence after conversion surgery, but not high resectability, may be required for obtaining a survival benefit in the modern era. Therefore, the surgical indication for conversion surgery should be carefully decided in a multidisciplinary meeting and should be relatively limited according to radiological findings as well as the CA19-9 level. van Veldhuisen et al 31 have suggested that in addition to CA19-9, other promising biomarkers, such as micro-RNAs and circulating tumor DNA, may more accurately predict treatment response in UR-PDAC. [41][42][43] In the near future, reliable surrogate markers for predicting resectability, early recurrence, and favorable prognosis should be explored.
Moreover, the optimal timing between initial treatment and surgical resection, an accurate method to evaluate tumor remission, and the type/duration of multimodal therapy are still under investigation.
Several prospective studies are now in progress. 35,36 Sustainable efforts will be required to prolong survival in patients with UR-PDAC.

Number of candidates for conversion surgery is now increasing
with the introduction of modern chemotherapy regimens; however, the actual clinical benefits of resection have not yet been fully investigated. Although conversion surgery can improve longterm survival in patients with UR-PDAC, the early recurrence rate should be recognized. There are still several problems to be resolved in this area, and prospective studies will be needed to explore the clinical benefit of conversion surgery. An appropriate surgical indication for achieving desirable outcomes can definitely provide further improved survival and early recurrence rates.
Therefore, novel biomarkers predicting resectability, OS and DFS should be investigated in the near future.

D I SCLOS U R E
Conflicts of Interest: Authors declare no conflicts of interest for this article.