Treatment of mNPC was a major challenge for radiotherapy physician. The skeleton was the most common site of distant metastasis in NPC, whereas the optimal therapeutic strategy has remained largely undefined. Our study showed that compared to PCT alone, combination of PCT and LLRT could improve OS and PFS in de novo mNPC with bone-only metastasis, especially in patients with limited metastatic bone sites (≤ 3). Moreover, LLRT and number of bone metastasis (≤ 3) were favorable independently factor for OS. For patients receiving PCT plus LLRT, there was no significant benefit from RT to metastatic bones in unselected patients. Subgroup analysis further confirmed that RT for metastatic bones only improve OS and PFS for mNPC with less than 3 metastatic bone lesions.
Local radiotherapy in mNPC is becoming hot-button issues. Rusthoven et al. reported that compared to chemotherapy alone, local radiotherapy combined with chemotherapy was associated with improved OS (median OS 21.4 vs 15.5 months, 5-year OS 28% vs 10%), resulting in a 39% reduction of death risk[13]. Similar findings were also found in some retrospective studies[14–20]. Up to now, there are only one multicenter phase 3 randomized clinical trial investigated the efficacy of LLRT in de novo mNPC. The trial demonstrated that PCT plus LLRT can significantly prolong OS of chemotherapy-sensitive mNPC patients. The 2-year survival rate was 76.4% in PCT plus LRRT group versus 54.5% in the PCT alone group ( HR = 0.42) [5]. As to mNPC with bone-only metastasis, Shen et al found that compared with chemotherapy or radiotherapy alone, combined chemoradiotherapy could significantly benefit patients with single bone metastasis (HR = 0.21). In additional, spinal metastasis and more than 3 bone metastasis sites are unfavorable elements for OS [21]. Consistent with above studies, our study also highlight the significance of LLRT in de novo mNPC with bone-only metastasis. Our study suggest that patients receiving PCT plus LLRT had over three times OS than PCT alone (45.0 months vs 13.5 months), and longer OS was remarkably observed in patients who had less than 3 bone ( median OS = 63.5 months). While our results were somewhat different from Shen’ study, our study suggested that less chemotherapy cycles (< 4 ), no locoregional radiation therapy and larger number of bone metastasis (> 3) were associated with worse OS. Taken together, LLRT should be considered after PCT in clinical practice of mNPC in future, especially for mNPC patients with bone-only metastasis. More randomized clinical trials are warranted.
For mNPC patients with bone-only metastasis, who will benefit from RT to metastatic bones has not yet been well characterized. Li et al reported that patients who received intensive local radiotherapy to bone lesions had longer OS (HR = 0.63) and PFS (HR = 0.80), and the post-treatment EBV DNA level and radical radiation dose were independent prognostic factors for OS [9]. However, our study indicated that for patients receiving PCT plus LLRT, the benefit from RT to metastatic bones was only seen in mNPC with less than 3 metastatic bone lesions, not in patients without selection. Of note, patients receiving PCT plus LLRT and RT to metastatic bones lives the longest survival. Therefore, for mNPC with less than 3 metastatic bone lesions, radical radiotherapy to metastatic bone would be strongly preferred as patients may have a long-term survival.
Since there is no consensus on who should receive LLRT and / or RT to metastatic sizes after first-line PCT, prognostic models and risk stratifications guiding the subdivisions of mNPC patients based on clinical features, therapeutic response and blood biomarkers had been widely researched to identify the optimal candidates [3, 14, 17, 19, 22–25], as well as in mNPC with bone-only metastasis[6, 9–11, 21, 26]. Li et al suggested that instead of high-risk mNPC, low-risk patients experienced significant survival benefits from definitive radiation therapy in addition to palliative chemotherapy (PCT) based on 5 prognostic factors, including response of metastases to chemotherapy, number of metastatic sites, liver metastasis, serum lactate dehydrogenase and posttreatment EBV-DNA [22]. Zou et al reported PCT combined with LLRT may benefit mNPC patients without liver involvement liver involvement [3]. Xu et al found that additional LRRT after PCT may only improve OS for oligometastatic patients, rather than polymetastatic disease [23]. Similar findings were also seen in other studies [15, 24]. With respect to the mNPC with bone-only metastasis, chen et al. develop a prognostic score and divided patients into low and high risk groups based on age, N, anemia, bone metastasis free interval, without radiotherapy to primary sites and without radiotherapy to first metastasis sites [26]. Sun et al. subdivide mNPC with bone-only metastasis based on EBV DNA after PCT and the number of metastatic lesions[10]. Our study classified mNPC into two group through number of bone metastatic lesions, proving a simple and feasible way to predict survival and guide clinical practice. In summary, looking for reliable and valuable biomarkers is vital to aid doctors in selecting the most suitable patients for individualized and comprehensive therapy.
The study had sone limitations. Firstly, our study is retrospective study in single-center. Secondly, our simple sizes are relatively small, which might affect statistical performance even when PSM was performed. Thirdly, EBV DNA and other blood biomarkers were not assessed in our study. Further prospective trails are needed in the future to guide the management of de novo mNPC with bone-only metastasis.