Molecular testing for personalized therapy is underutilized in patients with borderline resectable and locally advanced pancreatic cancer – real world data from the NORPACT-2 study

Abstract

Background: International guidelines currently recommend the use of molecular testing in patients with advanced pancreatic cancer. The rate of actionable molecular alterations is low. The utility of molecular testing in patients with borderline resectable (BRPC) or locally advanced (LAPC) pancreatic cancer in real world clinical practice is unclear.

Methods: 188 consecutive patients included in a prospective, population-based study (NORPACT-2) in patients with BRPC and LAPC (2018–2020) were reviewed. Molecular testing was performed at the discretion of the treating oncologist and was not recommended as a routine investigation by the national guidelines. All patients were considered fit to undergo primary chemotherapy and potential surgical resection. The frequency and the results of molecular testing (microsatellite instability (MSI) and/or KRAS status) were assessed.

Results: Thirty patients (16%) underwent molecular testing. MSI tumour was detected in one (3.6%) of 28 tested patients. The patient received immunotherapy and subsequently underwent surgical resection. Histological assessment of the resected specimen revealed a complete response. KRAS wild type was detected in one (14.3%) of seven tested patient. Patients who initiated FOLFIRINOX as the primary chemotherapy regimen (p = 0.022), or were being treated at one of the eight hospital trusts (p = 0.001) were more likely to undergo molecular testing.

Conclusions: Molecular testing was rarely performed in patients with BRPC or LAPC. Routine molecular testing for all patients with BRPC and LAPC should be considered to increase identification of targetable mutations and improve outcomes.

Keywords:

• pancreatic cancer
• molecular testing
• chemotherapy
• surgery
• KRAS
• microsatellite instability

Introduction

Pancreatic cancer remains a formidable challenge in oncology due to its aggressive nature and the limited treatment options. Without advancements in precision oncology, the prognosis will remain dismal [1]. Molecular testing may guide personalized treatment strategies in advanced pancreatic cancer. The National Comprehensive Cancer Network (NCCN) and European Society of Medical Oncology (ESMO) guidelines recommend the utilization of molecular analysis techniques in this patient population [2, 3]. However, the guidelines vary among Scandinavian countries regarding the recommendation for molecular profiling in patients with pancreatic cancer [4–6]. ESMO recommends KRAS and BRCA testing, and if a KRAS-wild type (wt) tumour is identified, additional analysis for the detection of microsatellite instability (MSI) status, NTRK fusion status and other rare, potentially actionable findings [3]. If multigene sequencing is not carried out, ESMO currently recommends that detection of MSI and NTRK fusions should be done using standard methods [3]. NCCN recommends MSI testing and/or MMR testing on available tumour tissue from patients with locally advanced (LAPC) or metastatic pancreatic cancer, and to consider specifically testing for actionable somatic findings [2]. In borderline resectable pancreatic cancer (BRPC), NCCN currently recommends core tumour biopsy to be performed at the time of diagnosis to obtain adequate tissue for possible ancillary studies [2].

Only a few studies have addressed the utility of molecular testing in real world clinical practice in patients with BRPC and LAPC [7]. The aim of the study was to assess the frequency and results of molecular testing in patients with BRPC and LAPC who were included in the Norwegian Pancreatic Cancer Trial-2 (NORPACT-2), a prospective, population-based cohort study [8]. Molecular testing (KRAS and/or MSI testing on tumour tissue) was neither part of the trial protocol nor specifically recommended in the national guidelines during the period of the trial. Accordingly, the current study evaluates how molecular testing is utilized when performed at the discretion of the treating oncologist at the patient’s local hospital.

Material and methods

Study population

From January 2018 to December 2020 all patients with BRPC or LAPC referred to Oslo University Hospital (OUH) from the South-Eastern Norway Regional Health Authority were offered participation in a prospective, observational cohort study (NORPACT-2), and included upon written informed consent. OUH covers a population of 3.1 million people (of a total population of 5.4 million) within a geographically defined area. The study protocol was approved by the Regional Ethical Committee, Norway (REC Nord 2017/1382). Biobanking of blood or tumour biopsies was approved by the Regional Ethics Committee (REK 2015/738). Diagnostic work-up, treatment and follow-up followed the national guidelines, as previously described [8, 9]. All patients were discussed at the multidisciplinary team meeting at OUH at the time of diagnosis and during response evaluations to consider the patients for surgical resection. All patients had a confirmed cytological or histological diagnosis of pancreatic cancer upon initiation of chemotherapy. Tissue diagnosis was obtained by endoscopic ultrasound-guided fine-needle aspiration/biopsy (EUS-FNA/FNB) [9]. MSI or KRAS analysis on available tumour samples was performed at Section of molecular pathology, Department of pathology, OUH. Oncological treatment, including chemotherapy, radiotherapy or potentially targeted therapy was given at the discretion of the treating oncologist at the patient’s local hospital. The different local hospitals within the South-Eastern Norway Regional Health Authority are organized in eight different hospital trusts.

Statistics

Statistical analyses were performed using SPSS version 29. Continuous variables were described as medians with interquartile ranges and compared using the Mann–Whitney test. Categorical variables were expressed as counts with % and compared using the χ² or Fisher’s exact test (for small numbers). Overall survival was measured from the time of diagnosis until death or the last follow-up (December 31, 2023). Crude differences between patient groups were assessed using the Kaplan–Meier method and the log-rank test. Statistical significance was set at p < 0.05.

Results

Of 230 patients, referred to and discussed at the multidisciplinary team meeting at OUH, 113 (49.1%) had BRPC and 117 (50.9%) LAPC. Forty-two patients (18.2%) received best supportive care only and were excluded from further analysis. Of the remaining 188 patients, 186 patients (98.9%) received primary chemotherapy, while two BRPC patients (1.1%) underwent upfront surgery followed by adjuvant chemotherapy. Overall 57 patients underwent surgical resection. During the course of the disease, 30 patients (16%) underwent molecular testing. Comparison of patients who did and did not undergo molecular testing is presented in Table 1. Patients who initiated FOLFIRINOX as primary chemotherapy regimen (p = 0.022), or were being treated at one of the eight hospital trusts (p = 0.001) were more likely to undergo molecular testing.

Table 1. Baseline and treatment characteristics for patients with borderline resectable or locally advanced pancreatic cancer fit to receive primary chemotherapy.

	Patients undergoing molecular testing (n = 30)	Patients not undergoing molecular testing (n = 158)	p-value
Age (median, IQR) (years)	66 (59–73)	70 (63–75)	0.163
Gender, n (%)			0.082
Male	20 (66.7)	78 (49.4)
Female	10 (33.3)	80 (50.6)
Performance status (ECOG), n (%)			0.187
0	22 (73.3)	88 (55.7)
1	7 (23.3)	57 (36.1)
>1	1 (3.3)	13 (8.2)
Charlson comorbidity index, n (%)			0.278
0	19 (63.3)	76 (48.1)
1	8 (26.7)	53 (33.5)
>1	3 (10)	29 (18.4)
Body mass index (median, IQR) (kg/m^2)	23.8 (22.8–77.2)	23.5 (20.9–26.5)	0.924
CA19-9 (median, IQR) (kU/l)	164 (75–657)	400 (74–1071)	0.109
Tumour classification, n (%)			0.186
Borderline resectable	12 (40)	84 (53.2)
Locally advanced	18 (60)	74 (46.8)
Tumour location, n (%)			0.832
Head/uncinate process	24 (80)	129 (81.6)
Body/tail	6 (20)	29 (18.4)
Primary chemotherapy regimen, n (%)*			0.022
(m)FOLFIRINOX	24 (80)	79 (50)
Gemcitabine-nab/paclitaxel or gemcitabine	6 (20)	69 (43.7)
Other	0 (0)	8 (5.3)
RECIST response at restaging before definitive decision of resectability^†			0.195
Complete/partial	3 (10)	15 (9.5)
Stable	19 (63.3)	102 (64.6)
Progression	8 (26.7)	35 (22.2)
Surgical resection, n (%)			0.409
Yes	11 (36.7)	46 (29.1)
No	19 (63.3)	112 (70.9)
Hospital trust, n (%)			0.001
1	12 (52.2)	11 (47.8)
2	4 (16.7)	20 (83.3)
3	5 (15.2)	28 (84.8)
4	4 (13.8)	25 (86.2)
5	3 (11.5)	23 (88.5)
6	2 (7.4)	25 (92.6)
7	0 (0)	9 (100)
8	0 (0)	17 (100)
Median overall survival (months, 95% CI)	21 (13.5–28.5)	14 (11.8–16.2)	0.156

Abbreviations: CI: confidence interval; ECOG: Eastern Cooperative Oncology Group; (m) FOLFIRINOX: (modified) 5-fluorouracil with leucovorin, irinotecan and oxaliplatin; IQR: interquartile range; RECIST: Response Evaluation Criteria In Solid Tumours.

*Two BRPC patients undergoing upfront surgery followed by adjuvant chemotherapy excluded from analysis.

†Two patients underwent upfront surgery, and two BRPC and two LAPC patients starting primary chemotherapy did not undergo the planned first restaging computed tomography.

The molecular testing results for the 30 patients are summarized in Table 2. Of the 28 patients tested for MSI, one (3.6%) was found to have a MSI tumour and diagnosed with Lynch syndrome, resulting in significant treatment implications. After experiencing toxicity on first and second line chemotherapy, the patient received immunotherapy for one year and subsequently underwent surgical resection. Histological assessment of the resected specimen revealed complete tumour regression. Among the seven patients tested for KRAS status, one (14.3%) was found to have wild-type KRAS, but further testing for actionable mutations was not conducted. The remaining six patients who were tested had KRAS mutation. Inadequate primary biopsy samples hindered successful molecular profiling in five patients. Repeat EUS-FNB in two patients and percutaneous or laparoscopic biopsy of metastatic lesions in three patients provided sufficient tumour tissue for molecular analysis (Table 2).

Table 2. Results of molecular testing in 30 patients with borderline resectable or locally advanced pancreatic cancer.

	n (%)
Microsatellite instability testing performed	28 (93.3)
Microsatellite stability	27
Microsatellite instability	1
KRAS testing performed	7 (23.3)
KRAS mutation	6
KRAS wild type	1
Sample used for molecular profiling
Endoscopic fine needle aspiration at time of diagnosis	12
Endoscopic fine needle biopsy at time of diagnosis	7
Resection specimen	6
Laparoscopic peritoneal biopsy	1
Percutaneous biopsy of paraaortal lymph node metastasis	1
Percutaneous biopsy of liver metastasis	1
Repeat endoscopic fine needle biopsy performed after initiation of chemotherapy	2
Initiating personalized treatment based on molecular testing	1 (3.3)
	Median, months (IQR)
Time from diagnosis to molecular testing	7 (3–26.4)
Time from molecular testing to death	10.5 (4–22.5)

Abbreviation: IQR: interquartile range.

Discussion

Opinions currently differ as to whether molecular testing provides relevant information for clinical decision-making in patients with BRPC or LAPC. Proponents argue that upfront molecular testing is always required in patients fit for oncological treatment in order to identify the few who may benefit from personalized therapy [10, 11]. The low rate of molecular testing in the current study (16%) emphasizes the need for increased awareness and approximation to the NCCN and ESMO guidelines, which recommend molecular testing in patients with BRPC and LAPC. The study reveals that based on the judgment of a single treating oncologist, molecular testing exceeded significantly the recommendations of the national guidelines at one of the eight hospital trusts. This highlights the variation in clinical practice as well as the need for standardization of molecular testing and the importance of clear national guidance on this issue in order to ensure that all patients will receive appropriate testing.

The underutilization of molecular analysis may represent a gap in patient care. Molecular routine testing in pancreatic cancer is crucial for identifying potential targets for personalized treatment [12]. The identification of an MSI tumour in one (3.6%) of 28 tested patients and KRAS wt in one (14.3%) of seven tested patients highlights the potential clinical impact of molecular analysis. Even though only few patients may have actionable aberrations that open up for targeted cancer therapy, it appears essential to test all patients to uncover those few. MSI/dMMR has a prevalence of around 1%–2% in pancreatic cancer [13]. MSI has been associated with increased responsiveness to immunotherapy, making it an important finding with implications for treatment decisions [14, 15]. KRAS wt pancreatic cancer accounts for up to 5%–7% of all cases and harbours potentially druggable alterations, including immune-oncological targets [16]. Identifying patients with KRAS wt tumours in routine clinical care could broaden treatment possibilities in a significant way [16]. Importantly, ongoing studies such as IMPRESS Norway provide patients with the opportunity for inclusion in experimental trials and a greater chance of undergoing extended molecular analysis if, for example, they have MSI or KRAS wt tumours [17]. By conducting routine molecular testing on pancreatic cancer at the time of diagnosis, potential genomic alterations can be detected. These alterations may make patients eligible for current clinical trials or support off label use of an agent approved for another indication [10, 17]. From a payer’s perspective, evidence supporting the cost-effectiveness of using multigene sequencing in routine practice is limited [18]. Therefore, it is not currently recommended to perform multigene next-generation sequencing in patients with advanced pancreatic cancer in daily clinical practice. However, ESMO advises using molecular screening programs with cheaper standard methods [18]. Aligning with this recommendation, the Norwegian guidelines were recently updated to recommend MSI and KRAS analysis at the time of diagnosis for patients who are suitable candidates for chemotherapy or potential targeted therapy.

In five patients, obtaining EUS-FNA/FNB samples that were adequate for molecular analysis proved to be challenging. Although tissue acquisition under EUS guidance is an established diagnostic method, procedural outcomes for molecular profiling have been variable [19]. In the future, the development and implementation of more advanced techniques for obtaining adequate tissue samples, such as liquid biopsies or novel endoscopic approaches, may overcome the challenges associated with EUS-FNA/FNB.

The vision for the future is that pancreatic cancer treatment will be biomarker driven and consequently, a growing array of targeted therapeutic options based on novel or existing biomarkers is emerging [20]. This emphasizes the importance of improving primary diagnostics, such as liquid biopsies or endoscopic biopsies from the primary tumour or metastases, to obtain adequate tissue for molecular testing. Given the increasing availability of targeted therapies and precision medicine approaches in pancreatic cancer, routine molecular testing should be considered for all patients with BRPC and LAPC patients fit to undergo medical or surgical treatment to guide treatment decisions and improve outcomes.

Acknowledgements

The authors acknowledge the financial support provided by grants from the Norwegian Cancer Society (grant numbers 198039-2018 and 212734-2019 (The Norwegian Cancer Society’s National Group of Expertise on Pancreatic Cancer Research)) and the South-Eastern Norway Regional Health Authority (grant numbers 2018088 and 2019029). The authors are grateful to the study nurses Asle Sandved Rudjord and Eline Angard Ulateig for their excellent work in the project office and with patients.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the South-Eastern Norway Regional Health Authority [grant numbers 2018088 and 2019029] and the Norwegian Cancer Society [grant number 198039-2018 and 212734-2019].