Do you electively treat the ipsilateral hilum in patients with peripheral limited stage small cell lung cancer who have negative pathological sampling of mediastinal and hilar nodes by EBUS and no FDG-avid nodes on PET?

These patients by your description essentially have stage I SCLC. They may be candidates for lobectomy followed by adjuvant chemo (or chemo followed by lobectomy, if surgical candidates).

Similarly while there are a range of acceptable approaches to these patients if they are not/marginal surgical candidates, our practice is to offer SBRT followed by adjuvant chemotherapy. 

This is supported in the literature with institutional and multi-institutional pooled reviews demonstrating excellent local control (96% at 3 years), and roughly 50% long-term DFS in this limited sample of 74 patients. 

The question of nodal failure after SBRT is valid in that almost a quarter of patients did experience nodal failure, though typically in the context of having other sites of recurrence disease (unclear whether lack of nodal coverage contributed to failure, or whether this was part of the manifestation of broader recurrence). Also, only a quarter of patients in this series had upfront mediastinal staging (likely due to limited availability of EBUS at the time of these patients treatment). In older surgical series there is substantial demonstration of occult nodal disease (for instance 11 of 36 patients in this series with clinical stage Ia and Ib disease). More modern surgical series are available, though limited by frequent lack of patterns of failure data so unfortunately not as informative on XRT field design as otherwise possible. Currently with routine availability of EBUS we expect more accurate clinical staging.

Delivery of chemotherapy (before or after SBRT) was associated with substantially improved OS in this series of stage I patients receiving SBRT, and one would expect this to have an impact upon occult nodal disease. Sequencing of chemo is variable, though our preference is SBRT prior to chemotherapy given the ability to rapidly deliver SBRT, the risk of complications of chemotherapy in a potentially medically frail population, and more clear definition of the pre-chemotherapy target volume. 

The delivery of PCI in stage I disease is controversial, particularly given the relatively lower incidence of brain metastasis in early stage patients and I do not favor PCI, though other colleagues would offer this even in stage I disease (23% of the pooled analysis patients).