Nitrates in drinking water and cancers of the colon and rectum: a meta-analysis of epidemiological studies

1

Introduction

The possible relationship between increased risks of cancers of the colon and rectum and nitrate ingestion from drinking water has emerged as a major environmental issue. Cancer of the colon and rectum, considered together as colorectal cancer, is the third most common cancer (after breast and lung) in both sexes worldwide, with 1.9 million new cases in 2020, and is second in terms of mortality [12] . Nitrate is a frequent drinking water pollutant, with a major source being agricultural activities, and in some countries nitrate levels in drinking water supplies have been increasing due to more intensive agriculture and livestock farming [3] .

The exposure limit for nitrate in drinking water set by the World Health Organisation (WHO) and adopted by many countries, is 50 mg/l NO ₃ (nitrate ion), the equivalent in NO ₃ -N (nitrogen) being 11.3 mg/l, conversion factor 0.226 [4] . The limit set in the US and Canada is approximately the same, 10.0 mg/L NO ₃ -N.

This level was set in the 1950 s to protect against methaemoglobinaemia, “blue baby syndrome”, which occurs mainly in infants and was linked to drinking water from shallow wells in the US with high levels of nitrates [56] . Daily intake of nitrates from an average food consumption and average water intake have been estimated at 56–86 mg/day [7] , of which about 90% comes from diet and 10% from water. However the proportion ingested from water can rise to over 50% if nitrate concentrations approach or exceed the WHO standard of 11.3 mg/L NO ₃ -N.

The relationship between nitrates and cancer risk is complex. In their major review published in 2010 [7] , the International Agency for Research on Cancer concluded that there was “inadequate evidence” in humans for the carcinogenicity of nitrate in drinking water, or of nitrate in food, and also “inadequate evidence” in animals for the carcinogenicity of nitrate. However, there was “sufficient evidence” in experimental animals for the carcinogenicity of nitrate in combination with amines or amides, and “limited evidence” in humans for the carcinogenicity of nitrites in food, based on an increased incidence of gastric cancer. The overall assessment was dependent on the acceptance of the mechanism that the carcinogenic effects of ingested nitrates are produced by the formation of nitroso compounds, and the production of these is increased by nitrosatable compounds found in meat, and inhibited by vitamin C and other antioxidants found in vegetables. Thus for the first time in its series of reviews, the IARC report assessed the carcinogenicity of nitrates and nitrites in terms of the assumed mechanism, and classified “ingested nitrate or nitrite under conditions that result in endogenous nitrosation” in group 2 A, “probably carcinogenic to humans”.

Since that report, further studies on drinking water nitrate and colorectal cancer have appeared. A retrospective cohort study based on the whole population of Denmark estimated drinking water exposures on the basis of residential histories up to age 35, and used cancer incidence from age 35 to a maximum of 68, with mean follow-up of 13.1 years [8] , finding a positive relationship with an estimated increase of 4% per 1 mg/l increase in drinking water nitrate, over a range from very low levels up to the maximum decile category, over 3.8 mg/l. The other cohort study, of postmenopausal women in Iowa, USA recruited in 1986, used exposure data from 1955 to 1988, and cancer incidence from 1986 to 2010 [9] , and found no increase in colon cancer up to a maximum category of over 3.5 mg/l, and a significant decrease in rectal cancer incidence at the higher nitrate levels. Several case-control studies are also available, with varying results. A study in Spain and Italy [10] recruited incident hospital cases in 2008–2013, and estimated drinking water nitrates from residential histories from 30 to 2 years before interview. In Wisconsin, USA, incident cases diagnosed in 1990–92 and 1999–2001 were identified, and nitrate data from groundwater and wells sampled in 1994 was used [11] . In Iowa, USA, cases incident in 1986–87 were identified and water exposures from 1960 onwards used [12] . Studies based on deaths from colon and rectal cancer in Taiwan, using residential histories for water exposures, were published in four papers 13141516 .

A meta-analysis published in 2019 [17] reported a positive relationship similar to the Danish study, which by its size dominated that analysis; but this meta-analysis did not include the Iowa cohort study, published later, and was restricted to colon cancer. Another meta-analysis including both cohort studies and reported that drinking water nitrate was not associated with colorectal cancer risk [18] , although it did find a positive, but not significant, trend. Another meta-analysis showed an association with colon cancer, but not with rectal cancer; but this analysis included one ecological study while omitting several others, and included two overlapping papers from the Iowa cohort study [19] .

For the current meta-analysis, all studies up to June 2021 are included, and both colon cancer and rectal cancer are considered.