Sir,

We read with interest the comments from Ji and Chen (2013) on our recent publication providing evidence for the potential use of β-blockers in the treatment of neuroblastoma (Pasquier et al, 2013). This letter raised some interesting points and we agree that the effects of β-blockers on tumour endothelial cells and the role of adrenergic signalling in the biology of neuroblastoma both deserve further investigation.

Our work on the use of β-blockers for cancer treatment aroused from the serendipitous discovery of the efficacy of propranolol in treating severe infantile haemangioma (Leaute-Labreze et al, 2008). This finding led us to hypothesise that β-blockers may be able to increase the efficacy of chemotherapy against drug-refractory cancers, either through direct effects on cancer cells and/or via anti-angiogenic mechanisms. To date, we have screened the anti-angiogenic, anti-proliferative and chemo-sensitising properties of β-blockers alone and in combination with more than 10 different chemotherapeutic drugs in 17 different cell lines, including endothelial (4), breast cancer (3) and neuroblastoma cells (6) (Pasquier et al, 2011, 2013; and personal unpublished observation). We found that some β-blockers (i.e., propranolol, carvedilol and nebivolol) were consistently able to (i) inhibit angiogenesis in vitro and (ii) potentiate the anti-angiogenic and anti-proliferative effects of chemotherapy agents at concentrations that were 5- to 50-fold lower than those required to inhibit the proliferation of cancer cells when used alone. This goes to show that β-blockers are potent anti-angiogenic and chemo-sensitising agents, rather than true cytotoxic drugs.

In their letter, Ji and Chen (2013) focus their attention on the fact that β-blockers alone were able to slow down the growth of neuroblastoma tumours in TH-MYCN mice in our study. Although statistically significant, this anti-tumour effect did not translate into prolonged survival, indicating that this effect was transient. Importantly, the combination of β-blockers with vincristine increased the median survival of TH-MYCN mice up to 4-fold as compared with vincristine alone. These results were consistent with those observed with the combination of propranolol and paclitaxel in an orthotopic xenograft model of triple-negative breast cancer (Pasquier et al, 2011). Altogether, our compelling preclinical data indicate that β-blockers are more likely to be beneficial when used in combination with chemotherapy, rather than used as single agents.

One concern raised by Ji and Chen (2013) was the use of relatively high doses of β-blockers in our in vivo experiments. The doses used in this study were well tolerated by the animals, thus highlighting the inherent difficulties of extrapolating between doses used in animal studies and those effective in the clinic. Elsewhere, it is important to note that the use of standard doses of β-blockers for the management of hypertension has been associated with clinical benefits in melanoma, breast, ovarian and non-small-cell lung cancer patients (Powe et al, 2010; Barron et al, 2011; De Giorgi et al, 2011; Lemeshow et al, 2011; Melhem-Bertrandt et al, 2011; Diaz et al, 2012; Wang et al, 2013). This clearly shows that β-blockers can be beneficial in cancer treatment, even when used at standard low concentrations. Higher doses may, however, be required to maximise the anti-cancer effects of β-blockers, which may lead to unwanted side effects such as hypotension and hypoglycaemia. Future clinical trials will therefore need to address the optimal dosage of β-blockers for a given clinical setting.

Finally, Ji and Chen (2013) highlighted a pertinent clinical perspective of our work. Recent clinical trials have focused on minimising treatment for patients with non-high-risk neuroblastoma and confirmed the excellent outcome of these patients (Hero et al, 2008; De Bernardi et al, 2009; Baker et al, 2010; Rubie et al, 2011; Strother et al, 2012). Many non-high-risk patients experience spontaneous tumour regression that may take up to 18 months to become evident (Hero et al, 2008). Unfavourable histopathology, tumour ploidy and genomics have emerged as important prognostic markers of increased risk of treatment failure, including in non-high-risk patients (Janoueix-Lerosey et al, 2009). These studies highlight the increasing importance of careful and refined risk stratification to determine the treatment approach for children with neuroblastoma. We caution against widespread use of β-blocker therapy in unselected, non-high-risk patients, most of whom have excellent clinical outcomes with minimal therapy. However, we advocate for the clinical investigation of the potential of β-blockers to (i) further reduce treatment intensity in selected non-high-risk patients with unfavourable prognostic features, (ii) improve outcome of non-high-risk patients who fail observation or recur after surgery and (iii) salvage patients with relapsed and/or drug-refractory neuroblastoma tumours. Furthermore, β-blockers may also prove useful in the context of cancer care in low- and middle-income countries, where multiple constraints call for the use of orally available, inexpensive drugs with low toxicity profiles (Andre et al, 2013).