Population-based analysis of treatment patterns and outcomes for pancreas cancer in Victoria

SCANPatient: study protocol for a multi-centre, batched, stepped wedge, comparative effectiveness, randomised clinical trial of synoptic reporting of computerised tomography (CT) scans assessing cancers of the pancreas

BACKGROUND

Complete surgical removal of pancreatic ductal adenocarcinoma (PDAC) is central to all curative treatment approaches for this aggressive disease, yet this is only possible in patients technically amenable to resection. Hence, an accurate assessment of whether patients are suitable for surgery is of paramount importance. The SCANPatient trial aims to test whether implementing a structured synoptic radiological report results in increased institutional accuracy in defining surgical resectability of non-metastatic PDAC.

METHODS

SCANPatient is a batched, stepped wedge, comparative effectiveness, cluster randomised clinical trial. The trial will be conducted at 33 Australian hospitals all of which hold regular multi-disciplinary team meetings (MDMs) to discuss newly diagnosed patients with PDAC. Each site is required to manage a minimum of 20 patients per year (across all stages). Hospitals will be randomised to begin synoptic reporting within a batched, stepped wedge design. Initially all hospitals will continue to use their current reporting method; within each batch, after each 6-month period, a randomly selected group of hospitals will commence using the synoptic reports, until all hospitals are using synoptic reporting. Each hospital will provide data from patients who (i) are aged 18 or older; (ii) have suspected PDAC and have an abdominal CT scan, and (iii) are presented at a participating MDM. Non-metastatic patients will be documented as one of the following categories: (1) locally advanced and surgically unresectable; (2) borderline resectable; or (3) anatomically clearly resectable (Note: Metastatic disease is treated as a separate category). Data collection will last for 36 months in each batch, and a total of 2400 patients will be included.

DISCUSSION

Better classifying patients with non-metastatic PDAC as having tumours that are either clearly resectable, borderline or locally advanced and unresectable may improve patient outcomes by optimising care and treatment planning. The borderline resectable group are a small but important cohort in whom surgery with curative intent may be considered; however, inconsistencies with definitions and an understanding of resectability status means these patients are often incorrectly classified and hence overlooked for curative options.

TRIAL REGISTRATION

The SCANPatient trial was registered on 17th May 2023 in the Australian New Zealand Clinical Trials Registry (ANZCTR) (ACTRN12623000508673).

Changing patterns of care for pancreas cancer in Victoria: the 2022 Pancreas Tumour Summit

BACKGROUND

The Victorian Government convened the second Pancreas Cancer Summit in 2021 to identify unwarranted variation in care 2016-2019, and to assess trends compared with the first Summit 2017 (reporting 2011-2015). State-wide administrative data were assessed at population level in alignment with optimal care pathways across all stages of the cancer care continuum.

METHODS

Data linkage performed by Centre for Victorian Data Linkage combined data from Victorian Cancer Registry with other administrative data sets including Victorian Admitted Episodes Dataset, Victorian Radiotherapy Minimum Data Set, Victorian Emergency Minimum Dataset and Victorian Death Index. A Cancer Service Performance Indicator audit was carried out providing an in-depth analysis of identified areas of interest.

RESULTS

Of 3138 Victorians diagnosed with pancreas ductal adenocarcinoma 2016-2019, 63% were metastatic at diagnosis. One-year survival increased between time periods, from 29.7% overall 2011-2015 (59.1% for non-metastatic, and 15.1% metastatic) to 32.5% overall 2016-2019 (P < 0.001), 61.2% non-metastatic (P = 0.008), 15.7% metastatic (P = NS). A higher proportion of non-metastatic patients progressed to surgery (35% vs. 31%, P = 0.020), and more received neoadjuvant therapy (16% vs. 4%, P < 0.001). Postoperative mortality following pancreatectomy at 30 and 90 days remained low at 2%. Utilization of 5FU-based chemotherapy regimens increased between 2016 and 2020. Multidisciplinary Meeting (MDM) presentation was still below the 85% target (74%) as was supportive care screening (39%, target 80%).

CONCLUSIONS

Surgical outcomes remain world-class and there has been an appropriate shift in chemotherapy administration towards neoadjuvant timing with increasing use of 5FU-based regimens. MDM presentation rates, supportive care and overall care coordination remain areas of deficiency.

Characteristics of patients diagnosed with pancreatic cancer who access palliative care: An observational study

PURPOSE

Despite the benefits of palliative care (PC) in pancreatic cancer, little is known about patients who access PC. This observational study examines the characteristics of patients with pancreatic cancer at their first episode of PC.

METHODS

First-time, specialist PC episodes captured through the Palliative Care Outcomes Collaboration (PCOC), in Victoria, Australia between 2014 and 2020, for pancreatic cancer, were identified. Multivariable logistic regression analyses examined the impact of patient- and service-level characteristics on symptom burden (measured through patient-reported outcome measures and clinician-rated scores) at first PC episode.

RESULTS

Of 2890 eligible episodes, 45% began when the patient was deteriorating and 32% ended in death. High fatigue and appetite-related distress were most common. Generally, increasing age, higher performance status and more recent year of diagnosis predicted lower symptom burden. No significant differences were noted between symptom burden of regional/remote versus major city dwellers; however, only 11% of episodes recorded the patient as a regional/remote resident. A greater proportion of first episodes for non-English-speaking patients began when the patient was unstable, deteriorating or terminal, ended in death and were more likely to be associated with high family/carer problems. Community PC setting predicted high symptom burden, with the exception of pain.

CONCLUSION

A large proportion of first-time specialist PC episodes in pancreatic cancer begin at a deteriorating phase and end in death, suggesting late access to PC. Timely referrals to community-based specialist PC, access in regional/remote areas, as well as development of culturally diverse support systems require further investigation.

The association between quality care and outcomes for a real-world population of Australian patients diagnosed with pancreatic cancer

BACKGROUND

This study: (i) assessed compliance with a consensus set of quality indicators (QIs) in pancreatic cancer (PC); and (ii) evaluated the association between compliance with these QIs and survival.

METHODS

Four years of data were collected for patients diagnosed with PC. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). A multivariable analysis tested the relationship between significant patient and hospital characteristics, patient cluster effects within hospitals and survival.

RESULTS

1061 patients were eligible for this study. Significant association with improved survival were: (i) in the potentially resectable group having adjuvant chemotherapy administered following surgery or a reason documented (HR, 0.29; 95 CI, 0.19-0.46); (ii) in the locally advanced group included having chemotherapy ± chemoradiation, or a reason documented for not undergoing treatment (HR, 0.38; 95 CI, 0.25-0.58); and (iii) in the metastatic disease group included having documented performance status at presentation (HR, 0.65; 95 CI, 0.47-0.89), being seen by an oncologist in the absence of treatment (HR, 0.48; 95 CI, 0.31-0.77), and disease management discussed at a multidisciplinary team meeting (HR, 0.79; 95 CI, 0.64-0.96).

CONCLUSION

Capture of a concise data set has enabled quality of care to be assessed.

Management and Outcomes of Pancreatic Cancer in French Real-World Clinical Practice

Background: Our objective was to describe real-world patterns of care and outcomes in pancreatic cancer. Methods: 912 patients diagnosed with pancreatic cancer from 2014 to 2017 were registered by the population-based cancer registry of Burgundy (France). Progression-free and net survival were estimated. Results: at diagnosis, 52% of tumors were associated with metastases. Among the 20% of patients fulfilling resectability criteria, half of those aged 75−84 years and none of those ≥85 years actually underwent resection. Age was not associated with 3-year observed survival in patients who underwent resection. Overall, 77% of patients aged <75 years, 55% of those aged 75−84 years and 8% of those ≥85 years received chemotherapy. Among patients who were offered chemotherapy, 73% of those aged ≥85 years refused. Chemotherapy toxicity was higher with Gemcitabine_Oxaliplatin/Gemcitabine_Abraxane and FOLFIRINOX than with Gemcitabine alone. Patients resected after induction FOLFIRINOX and those treated with adjuvant Gemcitabine presented the lowest risk of progression. Three-year net survival was 35% in patients with non-metastatic resectable tumors and under 10% for other patients. Conclusions: Only half of patients aged 75−84 years with a resectable tumor actually underwent resection. Two thirds of patients aged ≥85 years refused chemotherapy, thus underlining the need to expand geriatric assessments.