BACKGROUND In 2003, we reported an increased risk of retinoblastoma in children conceived by IVF between 1995 and 2002. However, population-based studies among children conceived by IVF did not find an elevated risk of retinoblastoma. METHODS From nationwide estimates of numbers of live births conceived by IVF ( n = 40 330), we estimated the expected numbers of patients with retinoblastoma conceived by IVF in the period 1995–2007. The observed number of retinoblastoma diagnoses in children conceived by IVF was obtained by questionnaires sent to the parents of children with retinoblastoma diagnosed between 1995 and 2005. For non-responders and patients diagnosed after 2005, information was available through the medical files, in which information on fertility treatment has been routinely recorded since 2000. The relative risk (RR) of retinoblastoma among children conceived by IVF was calculated for the total study period (1995–2007) and for the expanded study period (2002–2007). RESULTS Of all eligible patients with retinoblastoma ( n = 162) diagnosed in the period 1995–2007, seven were conceived by IVF. In the total study period (1995–2007) the risk was significantly elevated [RR = 2.54, 95% confidence interval (CI) = 1.02–5.23]. In the expanded study period (2002–2007), no significantly elevated risk (RR = 1.29, 95% CI = 0.16–4.66) was found. CONCLUSIONS We found a significantly increased risk of retinoblastoma in children conceived by IVF in the total study period 1995–2007. However, this increased risk was mostly based on the much stronger risk increase observed previously, for 1995–2002. Caution and awareness on the one hand and avoiding unnecessary worries on the other hand are important at this stage of our knowledge.