Earthworms perform a number of essential functions in soil and are ideal indicators of soil pollution and declines in soil quality. The effects of pollutants on earthworm biomarkers are often investigated to assess the risk of soil pollution. The aim of this review is to explore whether the currently known earthworm biomarkers can be used in practice to assess soil pollution risk. In this paper, we review four aspects of earthworm biomarkers based on studies published in the past two decades and especially in the most recent 10 years. 1) The role of earthworm biomarkers in soil pollution risk assessment was first discussed, leading to the conclusion that although earthworm biomarkers cannot replace conventional chemical analytical methods, they offer a good complement for the flaws of chemical analysis, and may provide more comprehensive and straightforward information about soil pollution risk. Therefore, earthworm biomarkers may be very useful for primary screens in large-scale soil pollution surveys. 2) The earthworms tested in previous studies represented many different species and all three ecological categories (epigeic, anecic, and endogeic), but most studies involved Eisenia fetida/andrei (epigeic). Earthworm biomarkers could be classified into different groups: physiological-biochemical, individual behavior, morphological, molecular genetic and immunological. However, most researchers have focused on the anti-oxidant defense system, metallothionein (MT), avoidance response, lysosomal fragility (neutral red retention assay), and the comet assay. In addition, the range of exposure medium, exposure time, other components in the experimental design, and differences in the results were critically reviewed. We concluded that earthworm biomarker studies have been carried out mainly at the laboratory scale, that many biomarkers have been screened, and that the response mechanisms of typical earthworm biomarkers to typical pollutants and to combined pollutants have been elucidated to some extent. However, in the future, studies should focus on the feasibility of using earthworm biomarkers in real soil conditions. 3) Published studies vary greatly in experimental design and results, including the range of biomarkers and the response level in earthworms, and therefore are not directly comparable. This fact may pose a challenge to inferring broadly reliable conclusion and might limit the practical applications of this research. To elucidate the problems, we discuss the varying responses of superoxide dismutase, an important component of the antioxidant enzyme system in different earthworm studies; the results varied by three orders of magnitude which is confusing because the change of physiological-biochemical level should not be so large even in spiked earthworms. As a result, standard protocols for earthworm biomarker assays may be needed to aid comparisons among studies. 4) We recommend a set of research guidelines (such as the earthworm species to be tested, exposure times, types of biomarkers, etc.) based on the current studies. In addition, future studies should focus on methods to screen earthworm biomarkers that are simple, sensitive, and highly repeatable. Establishing a database of basal biomarker levels in earthworms in different soil types that is integrated with a national survey of pollution sources should be useful for practical applications in China. This review provided a global evaluation of earthworm biomarker studies and can serve as a reference for establishing standard earthworm biomarker assay protocols. Therefore, it will promote further studies and practical applications of earthworm biomarkers in assessing soil pollution.