Second Primary Cancers (SPC): A Mixed-Methods Systematic Review of Pathways to Diagnosis and a National Linkage Study in Scotland to Understand Survival Outcomes Following an SPC Diagnosis

Abstract

Background: As an increasing proportion of patients survive an initial cancer, the number who develop a second primary cancer (SPC) is also increasing. Aim: To examine the pathways to diagnosis of an SPC and the associated patients' and health practitioners' experiences, and survival outcomes following an SPC diagnosis. Methods: We conducted a systematic review to examine the pathways to diagnosis and associated patient and healthcare practitioners' experiences. Ten electronic databases were searched with no date or language restrictions. Two reviewers independently screened titles, abstracts and full-texts, and did the data extraction and quality assessment of the included studies. Secondary data analysis examined the survival outcomes following an SPC diagnosis. This involved three linked databases: the Scottish Cancer Registry, episode level data on hospital inpatient and day case discharges, and the National Register of Scotland death records. We included patients aged 18+, diagnosed with a first primary cancer (FPC) between 01.01.1980 and 31.12.2010 and subsequently diagnosed with an SPC (> 60 days following a FPC). The cases were matched for age, gender and cancer type to controls (individuals diagnosed with an FPC only within the same 5-year period). Descriptive statistics, survival analysis and Cox proportional hazard model were done. Results: Forty-nine articles were included in the review. The most commonly reported SPCs were breast, lung, colorectal, and melanoma. All included studies reported on the mode of detection of the SPC (30 symptomatic presentation, 18 screen detected, 34 other modes). The routes to detection (whether the SPC was diagnosed during routine follow-up/surveillance, or self-referral) varied by cancer type. Only two studies described provider experiences of diagnosis of an SPC; no study reported on the patients' experiences. The secondary data analysis included 37,193 people diagnosed with an SPC (51.6% females and 48.3% male). The proportion of SPC among people diagnosed with a FPC within a 5-year period ranged between 2.3% (1980-1985) and 31.2% (2006-2010). Trachea, bronchus and lung cancer (21.3%), colorectal cancer (13.9%), breast (12.9%) and prostate (7.3%) were the most frequently diagnosed SPCs in this population. These four cancers also accounted for the highest number of deaths in both sexes. Compared with people diagnosed with FPC (390 days: IQR 14-1825 days), those diagnosed with SPC had shorter median survival days (379 days: IQR 17-1825 days), which was statistically significant ( P = 0.001, X24df=163.2). Conclusion: The review provides early insights into diagnosis of an SPC. While it has identified that there is a dearth of quantitative and qualitative evidence on the pathways to diagnosis of an SPC, the secondary data analysis partly fills the gap, and has important implications for surveillance in both primary and secondary care, and secondary prevention and early detection.