Has time come for screening for occult cancer in patients with venous thromboembolism?

Abstract

The association between symptomatic venous thromboembolism (VTE) and the presence of a still asymptomatic malignancy has fascinated the medical community for almost 70 years. Over and over again the name of Armand Trousseau is quoted when reference is made to this association. However, upon careful reading of the original Trousseau lessons nothing is mentioned about occult cancer [1, 2]. In fact, he describes the other well-known association: patients with overt cancer often suffer from (recurrent) episodes of deep venous thrombosis, as well as superficial phlebitis. Two British physicians, James and Matheson, have the honour to be the first ones. They reported in 1935 the subsequent development of cancer in a patient who experienced a venous thrombosis 2 months earlier and speculated about the role of occult cancer as a possible cause of the thrombosis [3]. Subsequent retrospective and prospective studies have confirmed that the incidence of new cancers in patients with idiopathic VTE in the next 2–3 years is approximately 7.5%, while in those with secondary VTE this rate is around 1.6%[4-9]. There is also, in part from these studies, suggestive evidence that when these cancers are detected earlier they are predominantly at an early stage. The major challenge is whether we can diagnose these occult cancers and whether if detected the balance between good and harm is positive. In this issue of the Journal, Monreal and colleagues [10] present their interesting findings of a two-step approach to detect asymptomatic malignancies in a cohort of VTE patients. They included 1089 consecutive patients with a first episode of VTE (80% had deep vein thrombosis). At entry 17% had known cancer and they were excluded. The remaining 864 patients (40% with idiopathic VTE) underwent what the authors called a routine examination, including a thorough history taking, physical examination and laboratory investigation. Upon careful reading it is obvious that this ‘routine examination’ was quite extensive, with clear guidelines about subsequent investigations based on the findings. The relevance of this approach is that in less dedicated centers this examination may be less thorough and therefore less effective. Nevertheless, the authors identified 167 patients (19%) who based on this routine examination had a suspicion of an occult cancer, the majority (58%) based on careful history taking. The presence of cancer was confirmed in 34 patients, i.e. in a fifth of those suspected of having cancer, and 3.9% of all patients investigated. The remaining patients, i.e. those with a normal routine examination, as well as those with normal findings despite the initial suspicion of cancer raised during routine examination, underwent what the authors called a ‘limited diagnostic work-up’. This consisted of ultrasonography of the abdomino-pelvic region and several cancer laboratory markers. Of the 830 patients who underwent this testing, 13 (1.6%) were diagnosed as having cancer of which 60% were early stage. During the 1-year follow-up an additional 14 (1.7%) cases of cancer became apparent. Which questions are answered by this study and which are raised? The most impressive finding is the high yield of the initial, albeit quite standardized routine examination. Thirty-four of the total of 61 cancers seen in this study with a follow-up of 1 year were detected with this initial step. This is 56% of all cancers. The limited diagnostic work-up found another 21%, whereas 23% were missed. The second conclusion therefore is that the limited diagnostic work-up strategy needs to be improved. Other studies partly by the same investigators suggested that spiral computed tomography or other imaging modalities might increase the sensitivity of this step to above 90%[10]. Finally, although this study provides additional evidence that those cancers detected by screening are often early stage, we lack evidence that the overall benefit–risk ratio is positive. All investigators worldwide interested in this topic should cooperate in answering this question, only than time has come to include screening for occult cancer in patients with idiopathic VTE.