Should salvage surgery be considered for local recurrence after definitive chemoradiation in locally advanced non-small cell lung cancer?

The rates of local NSCLC recurrence after definitive CRT are even as high as 85 %
at 1 year 17], 18]. Pain, dyspnoea, haemoptysis and cough are frequent manifestations 20]. The local NSCLC relapse remains the leading death cause after the initial therapy
3].

The difficulties in distinguishing between tumor recurrence and fibrosis or inflammatory
changes on the computed tomography (CT) and FDG-PET are reported 24], 25, 26]. Dense consolidations are typical for radiation pneumonitis or fibrosis. However,
the tumor re-growth must also be suspected and considered differential diagnosis 2]. The increased SUVmax values should require further clearing by CT-guided or trans-bronchial
biopsy. On the other side, the excessive fibrosis or tumor necrosis has to be expected.
The fine-needle aspiration biopsy provides limited pathologic information. Therefore,
distinguishing between radiation- and tumor-induced fibrosis may lead to false-negative
findings. In some patients without tumor recurrence, a moderate FDG hypermetabolic
activity may persist up to 2 years after CRT 22]. Therefore, the optimal treatment strategy, when the histology is unclear, remains
undefined and suspected recurrence requires at least a comprehensive evaluation 21]. The right time window for the tumor biopsy and for beginning the therapy is crucial.
If only focal tumor nest surrounded by necrotic mass is evident, the further course
of the “remaining tumor” is unpredictable. Presence of the tumor cell nest must not
indicate the recurrence as those cells may become necrotic 19], 20]. On the other side, awaiting the radiological obvious relapse significantly decreases
the overall and the progressive-free survival time 14], 15]. Particularly, in patients with extended relapse invading the neighbouring structures,
the delay in decision making decreases the chance of the tumor resectability. The
biological nature of the local NSCLC relapse is less malignant in comparison to the
distant recurrence. Therefore the effective local treatment is likely to be associated
with prolonged overall survival and long-term disease control. In addition, some authors
recommend the salvage resections immediately after detecting the local recurrence
2].

The choice of a re-treatment strategy for local recurrence after the definitive CRT
remains challenging. Only few therapeutic alternatives have been reported. A systematic
literature review regarding the second line chemotherapy efficacy for relapsed NSCLC
shows a moderate symptom control and response rates for several anticancer drugs.
The median long-term survival and progressive-free survival remain under 9 and 6 months
respectively 21]. In addition, Hanna et al. report only a minimal increase in survival following re-chemotherapy
22]. The reported median long-term survival after reirradiation ranges from 5 to 14 months,
whereas the 2-year survival rates and the 2-year progression-free survival were 27 %
and 21 % respectively 17], 20]. Radiation tolerance of the lung, esophagus and the spinal cord limit the radiation
dose that could be applied to the recurrent tumor. Particularly, the myelopathy and
the late lung fibrosis have to be avoided. Jeremic et al. reviewed 11 studies of conventionally
fractionated external beam reirradiation for recurrent NSCLC. The analysis showed
improved overall survival after higher doses compared with low-dose reirradiation.
However, this was associated with an increased rate of pneumonitis and esophagitis
3]. Toxicity strongly results from applied total and cumulative median radiation dose.
However, locoregional and distant tumor control remains disappointing 22]. A radiofrequency ablation and cryoablation of the local recurrence is limited by
the position of the lesion and the tumor size. The percutaneous cryoablation is recommended
for more central lesions, in proximity to the large vessels 5]. The radiofrequency ablation is usually performed for lesion smaller than 3 cm. The
long-term local control is dependent on the tumor size and remains, particularly in
cases with lesion greater than 3 cm, unfavourable. Due to quick tumor progression
frequently repeated invasive procedures are required 6].

Recently published reports tend to recommend salvage lung resections as feasible treatment
option in absence of other management strategies for recurrent NSCLC. Due to limited
experience, the patient selection criteria for “post-radiotherapy” salvage lung surgery
are not clearly defined. Low postoperative mortality and complication rates despite
high-dose stereotactic body radiation support the idea of salvage surgery for local
recurrence in early stage NSCLC 10], 11]. Four other reports suggest that salvage surgery is associated with prolonged survival
in patients with locally recurrent or persistent tumor after definitive CRT in locally
advanced NSCLC 13]–16]. Bauman et al. state that the risk of salvage resection is proportional to the intensity
of fibrotic response after the high-dose radiation and to the interval between radiation
and resection. However, even in high-risk patients the salvage surgery was technically
feasible, with reasonable results, also when performed after long-time interval 14]. In addition, some authors identified salvage lung surgery as the best option for
patients with local tumor relapse, resulting in a prolonged survival 13], 23].

Kuzmik et al. reviewed 14 patients who completed definitive chemoradiation with median
dose of 57 Gy. After median interval between chemoradiation and surgery of 33 months
[range 0–169 months] local recurrence was identified in 54 % of the cases, locoregional
in 15 % and distant in 31 %. Viable tumor was found in all cases (100 %). The median
postoperative survival was 9 month with the 2-year survival rate of 49 % 13].

Bauman et al. described 24 patients treated initially with definitive chemoradiation
with a median radiation dose of 63.9 Gy. The median time between radiation and surgery
was 21 weeks. Nineteen patients (78 %) had pathological proof of viable tumor cells
in the resection specimen. The median survival time for entire cohort was 30 months
with estimated 3-year survival of 47 % 14].

Yang et al. described a cohort of 31 patients with various indications for salvage
resections. The chemoradiation was performed in 90 % of the cases, with median radiation
dose of 60Gy. The median interval between radiation and surgery was 17.7 weeks. Viable
tumor cells were identified in 19 (61 %) specimens. The median survival time for entire
group was 32.5 months with 3- and 5-year survival rates of 42 % and 31 %, respectively.
The patients with residual disease expected a median survival time of 20 months. Median
survival time in patients with complete pathologic response was 60 months and differed
statistically significant (p?=?0.03) 15].

Casirhagi et al. described an extended salvage resection in a group of 24 patients.
The median time between the definitive chemoradiation (mean 51 Gy) was 12 weeks and
was therefore close to the time interval for surgical resection after the induction
therapy. Viable tumor was found in 90 % of the cases. The postoperative 3-years survival
was 42 % 16].

Detailed study characteristics are summarized in the Table 4.

Table 4. Overview to available results for salvage lung surgery after definitive chemoradiation
in locally advanced NSCLC

Our group included only those patients with uncontrolled local disease (local relapse
or residual tumor following definitive CRT). The median disease-free survival in our
series was similar to the Bauman’s group (6.7 vs. 5 months). In contrast, Kuzmik et
al. performed salvage resections in patients with tumor recurrence in the contralateral
lung (31 %) and in the other ipsilateral lobe (15 %). Only 54 % out of 14 salvage
lung resections were performed for local recurrence. The median disease-free survival
for all patients was 33 months 13]. The ratio of extended salvage resections involving the neighbouring organs was in
our group higher (21 % by Bauman vs. 21 % by Kuzmik vs. 56 % Yang and 78 % our group).
Remarkable, despite the described differences the median long-term survival and estimated
3-year survival rates in our group were comparable with other series.

Majority of our patients developed locoregional (25 %) or distant recurrence (50 %)
during the follow-up interval. Therefore the long-term disease control remains crucial
and interdisciplinary decision making regarding further treatment is essential. Some
important points of agreement are as follow: 1) the salvage lung surgery should be
performed in patients with no other treatment alternatives; 2) the experience in this
new field in the thoracic surgery remains limited; 3) a careful patient selection,
particularly for extended resections, is fundamental.

Our study has a number of limitations. Small number of salvage resections was performed
and the data was collected retrospectively. A control group of patients with local
NSCLC recurrence after definitive CRT managed without salvage surgery was not available
for the comparison.

Salvage extended thoracic surgery may represent a good therapeutic alternative in
well selected cases with adequate pulmonary reserve and good performance status. The
individual patient targeted approach is essential and all alternative treatment options
should be discussed interdisciplinary.