NSCLC with PD is an advanced-stage type with poor prognosis[10]. The treatment exploring of those patients with PD has been still ongoing. Previous studies have elucidated the essential role of surgical resection in patients with PD[11]. Chen, et al confirmed the effectiveness of wait-and-see strategy for patients with PD[12]. Some preclinical study demonstrated the treatment value of near infrared photoimmunotherapy[13]. In conclusion, PD is a special type of advanced lung cancer, and its optimal treatment still needs further exploration.
Targeted therapy is the best therapeutic modality in NSCLC with PD. If there is corresponding gene mutation, the first choice is targeted therapy, which can greatly improve PFS and OS. For patients with EGFR activation mutation or ALK rearrangement, the use of tyrosine kinase inhibitors has completely changed the treatment of advanced NSCLC[14]. In our study, the 5-year OS rate and PFS rate of patients with PD was 48.1% and 20.5%, which was better than the prognosis of patients with other types of advanced lung cancer[15]. Meanwhile, the 5-year OS rate and PFS rate of patients without targeted therapy was 23.8% and 6.2%, which was also prior to the prognosis of patients with general advanced lung cancer. PD is a special type of metastasis causing the advanced staging, which differs from the metastatic type of lymph node metastasis or distant metastasis, such as brain or bone metastasis, whose prognosis is different from that of patients with other metastatic types. That is to say, NSCLC patients with PD has independent clinicopathological characteristics and prognosis.
Resection of the primary tumor focus is not helpful to improve the prognosis. In our institute, compared with simple biopsy of metastatic nodules, the resection of primary tumor, no matter wedge resection or lobectomy, has no effect on better PFS and OS. We speculate that although the primary lesion has been removed, there are still secondary metastatic lesions in the thoracic cavity, and there is still a persistent risk of progression or metastasis, which will affect the prognosis. Previous published articles demonstrated the potential surgical value of NSCLC with PD in selected patients, but only limited to patients without mediastinal lymph node metastasis[16, 17]. Li, et al. conducted a retrospective study enrolling 105 patients and revealed that surgery significantly prolonged survival in the non-targeted therapy group (median survival time 39.8 vs 14.2 months, P = 0.002)[18].The most accepted surgical modes were lobectomy and segmentectomy. Li, et al. enrolled 43 patients and found that patients in the tumor resection group had better PFS (3-year survival: 44.5% vs 0%; P = 0.009) and OS (3-year survival: 82.9% vs 38.5%; P = 0.013) than the open-close group did. However, there was no significant survival difference between sublobar resection and lobectomy for the main tumor resection[19]. Fan, et al obtained a similar conclusion[20]. Another retrospective study conducted by Li, et al. showed the prognostic benefit of surgery except for those with pericardial effusion or N3 disease[21]. Furthermore, patients who underwent lobe/bilobectomy had significantly better OS than those who underwent sublobar resection[21]. However, our study didn’t find the survival benefit from surgery, no matter PFS or OS, possibly due to the excellent prognostic contribution from targeted therapy and the small sample size of those without targeted therapy.
PCIC seems to help improve PFS. Inspite of the survival analysis of 58 patients or the survival analysis of patients who did not receive targeted drug therapy, two survival curves of whether receiving PCIC or not were obviously separated. However, patients might only benefit a limited degree from PCIC. In addition, our data volume is exactly small, so we failed to obtain statistically significant benefits. In 1995, Matsuzaki, et al proved the potential prognostic value of intrapleural perfusion hyperthermo-chemotherapy for patients with pleural dissemination who have had their primary lesions removed[22]. Moreover, PCIC is a safety, effective and promising approach[23] with controllable side effects[24]. At the same time, the length of hospitalization between two groups was not different. So PCIC could be considered in patients with occasional PD during operation or postoperatively.
Lymph node sampling or dissection was not helpful for the prognosis of patients with pleural cavity dissemination, and was even associated with worse PFS in univariate survival analysis. Multivariate analysis also did not find the impact of lymph node or sampling or dissection on the prognosis, which indicates that lymph node sampling or dissection is not necessary in patients with PD. In addition, univariate analysis showed that lymph node metastasis was associated with worse PFS, and had no significant effect on OS. Previous study showed that lymph node involvement was an independent prognostic factor for M1a patients[25]. Earlier in 1994, Akaogi, et al revealed that the presence of N2 disease was a poor prognostic sign in lung cancer with PD[26]. Iida, et al conducted a national study and found that better node status and macroscopic complete resection were beneficial to patients with pleural carcinomatosis[27]. PD is only a type of metastasis while there is still a possibility of no lymph node involved affecting prognosis in another way[28]. In my study, when the proportion of lymph node metastasis is lower, the prognosis is better. We wouldn't say that the 0.11 cutoff value is a referential indicator, but the lower the ratio of lymph node metastasis was, the better survival the patients had.
Our research has many limitations. First of all, this is a retrospective clinical study with many potential biases. In addition, the number of our samples is small. Although we have collected data from a single center for 8 years, only 58 patients were finally included after screening, and not all patients had lymph node sampling or dissection, which is not conducive to the selection of prognostic factors. In future, we can design a prospective, multicenter clinical study to determine the clinical interventions that affect the prognosis, such as PCIC.