Relative survival in early-stage cancers in the Netherlands: a population-based study

To the Editor,

Recently, Marcadis et al. performed a comprehensive analysis of relative survival in 281,970 patients with ten early-stage cancers using data from the Surveillance, Epidemiology, and End Results Program—a study that was hitherto lacking in contemporary literature [1]. They identified five early-stage cancers—i.e., ductal carcinoma in situ (DCIS) and early-stage cancers of the prostate, skin (melanoma), thyroid, and breast—with 10-year relative survival exceeding 100%, suggesting that patients with these early-stage cancers have greater longevity compared to an equivalent group from the general population. This finding favors the premise that these early-stage cancers—which presumably lack metastatic potential and have limited lethal consequences—are generally detected incidentally or via screening in individuals who are otherwise more healthy or health-conscious than their counterparts in the general population [2]. This phenomenon can to some extent be referred to as overdiagnosis [3].

At present, it is unknown whether the findings of Marcadis et al. extend to populations outside the USA. Therefore, we sought to complement their findings by delineating the relative survival of patients with early-stage cancers in the Netherlands.

We selected patients with ten early-stage cancers diagnosed during 2004–2015 from the nationwide Netherlands Cancer Registry (NCR). The selection of early-stage cancers and the study period is congruent with the selection as per Marcadis et al. [1]. The Privacy Review Board of the NCR approved the use of anonymous data for this study.

The primary endpoint was 10-year relative survival, defined as the time from diagnosis until death, emigration, or end of follow-up (February 1, 2019), whichever occurred first. Relative survival was calculated according to the cohort methodology to estimate disease-specific survival [4]. Relative survival is the ratio of the observed survival of patients to the expected survival of an age-, sex-, and calendar year-matched group from the general population. Expected survival was calculated as per the Ederer II methodology using Dutch life tables, stratified by age, sex, and calendar year. All analyses were performed using STATA (version 14.2, StataCorp).

Our analytical cohort included 225,305 patients with ten early-stage cancers diagnosed in the Netherlands during 2004–2015 (Fig. 1). Ten-year relative survival marginally exceeded 100% in patients with early-stage prostate cancer (101.4%; 95% CI, 100.7–102.0%), while it was close to 100% for patients with DCIS and stage I cancers of the skin (melanoma), testis, and thyroid (Fig. 2). Patients with stage I breast cancer experienced minimal excess mortality, while patients with early-stage oral/pharyngeal, bladder, lung, and pancreatic cancers experienced considerable excess mortality (Fig. 2).

In this nationwide, population-based study, the life expectancy of patients with early-stage prostate cancer slightly surpassed that of an equivalent group from the general population. This finding is congruent—albeit to a lesser extent—with those observed in the USA (Fig. 2) [1]. Besides, relative survival exceeded 100% among US patients with DCIS and early-stage cancers of the thyroid, skin (melanoma), and breast. In the Netherlands, however, relative survival in these patients ranges between 96% and 100%, which closely mirrors the disease-specific survival in the USA [1] Hence, relative survival in these early-stage cancers is likely to be overestimated in the USA. Seemingly, there is an overrepresentation of health-conscious individuals (e.g., those from higher socioeconomic groups or with adequate insurance coverage) in the US patient population with early-stage cancers. Moreover, these patients have a higher life expectancy than their counterparts in the general population and perhaps more diligently participate in periodic health examinations and cancer screening programs (i.e., healthy-user bias)—all of which may render a cancer diagnosis earlier in the disease course (i.e., lead-time bias) [2, 5, 6]. Sex- and treatment-related factors might also be at play with regard to disparities in relative survival between the Netherlands and the USA. The effect of sex and variation in treatment on relative survival could not be evaluated since information on sex and therapy was not analyzed in the study by Marcadis et al. [1].

Collectively, it can be reasoned that the magnitude of potential cancer overdiagnosis and health inequalities regarding access to screening, diagnosis, treatment, and follow-up services is less extensive in the Netherlands than in the USA, because all Dutch residents have equal access to health care services—regardless of their socioeconomic position and race, ethnicity, and gender. This topic provides an avenue for future research to close the gap in cancer survival between the Netherlands and the USA. This research requires detailed analyses on the drivers of survival disparities in early-stage cancers between the Netherlands and the USA.