**3.2 Surgery followed by radiotherapy**

The value of postoperative radiotherapy (PORT) for advanced head & neck cancers, was established in 1970's with few well researched studies [11]. Marcus et al. suggested doses for OC cancers of 6500Rads to achieve high local control and up to 7000rads for those with positive surgical margins [12].

Outcomes in locally advanced cancers are suboptimal with primary failure being loco-regional relapses. Hence combined modality treatments have been employed to counter this aspect as well as to ascertain if they added to overall survival patterns too. Lavaf et al. retrospectively analysed data from the Surveillance Epidemiology End Results (SEER) data base to collate results on 8795 lymph node positive cases of advanced head and neck cancers [13]. They found addition of adjuvant radiotherapy in these patients improved 5 yr. overall survival rates (43.2% vs. 33.4%; p<0.001) and cause specific survival rates (50.9% vs. 42.1%).

Kao et al. did a similar SEER group analysis of advanced cancers of head and neck region and specifically tried to address benefit of RT with respect to various nodal stages and specific sites of disease [14]. Subset analysis found that all nodal stages, including N1 Disease, has improved survival with the addition of adjuvant RT. Also, in multivariate analysis, addition of RT in node positive oral cavity tumours improved overall survival [HR, 0.84; 95% CI, 0.73–0.98; p = 0.025], though it was not as significant as compared to the other head and neck sites.

Another population-based study done by Shrimeet al, tried to specifically address the benefit of PORT in early-stage primary oral cavity cancers (pT1/T2N1) with single ipsilateral node without extracapsular spread and being <3 cm in size [15]. For all patients, adjuvant RT was associated with superior 5 yr. overall survival [52.4% vs. 41.4%(p<.001)]. Survival advantage was statistically significant in T2 cases, with a trend to significance in T1 cases. When individual sub-sites were analysed, PORT significantly improved survival in patients with cancers of oral tongue [52.3% vs. 37.9%(p=.002)] and floor of mouth [39.9% vs. 17.7%(p=.003)].

Addition of Chemotherapy to Radiation in post-op setting: Several retrospective and small prospective trials had showed improvement in outcomes of advanced head and neck cancers by adding chemotherapy (adjuvant or concurrent) to postoperative radiotherapy, especially in those with high-risk features such as inadequate margins, presence of extracapsular nodal spread and multiple involved nodes. Single institution trials by Bachaud and Smid, revealed improved loco-regional control rates and superior DFS/OS rates in combined modality group at a slightly higher rate of complications [16, 17].

Lacas & Pignon et al. (MACH-NC) in their updated meta-analysis of chemotherapy in head and neck cancer have studied the effect of concurrent chemotherapy use along with radiotherapy (upfront & post-surgery) [18]. Out of 107 studies (19,805 patients), 71 studies (10,680 patients) were on concomitant chemotherapy with a median follow up of 9.2 years. There was an absolute benefit in overall survival & event free survival of 6.5% & 5.8% at 5 years and 3.6% & 3.1% at 10 years with significant reduction in LRF rates [sub-HR = 0.71; p < 0.0001]. The study also showed that concurrent chemotherapy outcomes were better than induction chemotherapy and also that platin based chemotherapy gave maximum benefit, however there was no significant difference between weekly or three-weekly chemotherapy protocols in terms of outcomes or toxicity.

Bernier et al. (EORTC 22931) conducted a multi-institutional prospective study in 334 patients and found that addition of chemotherapy to adjuvant radiotherapy showed a benefit in 5 yr. PFS rates [47%(CMT) vs. 36%(RT)] and OS rates [53% (CMT) vs. 40%(RT)] & reduced the 5-year risk of death in combined modality from 43–27% [19]. Cooper et al.(RTOG 9501) did a similar study in 459 patients and showed significantly improved local/regional control as well as disease free survival (10% absolute benefit at 2 years) in the combined modality group after 45.9 months follow up [20].

As both these studies had few differences in their definition of high-risk disease, a pooled analysis was done to get some clarity and this concluded that outcomes for patients with ECE and/or involved surgical margins was significantly better with CMT [21]. Also, there was a trend to improved outcomes in patients with clinically enlarged level IV/V nodes with oral cavity/oropharyngeal primaries and perineural infiltration.

Three weekly high dose cisplatin (100 mg/m<sup>2</sup> ) is the standard systemic chemotherapy regimen given concurrently with radiation in high-risk oral cavity cancers. However, due to compliance issues low dose regimens have been applied in clinical practice without any prospective evidence to support the same. Szturz et al. conducted a meta-analysis to compare the standard three weekly regimen (100 mg/ m2 ; 3 doses) with the low dose weekly regimen ≤50 mg/m<sup>2</sup> ; ≥6 doses) [22]. Though there were no prospective comparative studies till date, 52 studies with 4209 patients were included and it showed no difference in the efficacy indices such as overall survival or response rates. In the definitive treatment setting, weekly regimen was more compliant and significantly less toxic with respect to myelosuppression, severe nausea/vomiting and nephrotoxicity. In the post-op setting, the two approaches were similar with the weekly arm causing more grade 3/4 dysphagia and weight loss.

JCOG1008 was a multi-institutional phase II/III non-inferiority trial comparing 3 weekly cisplatin with weekly schedule (40 mg/m<sup>2</sup> ) as a concurrent chemotherapy option in post-operative high-risk H&N cancers [23]. After enrolling 261 patients and a median follow up of 2.2 years, the trial was stopped as the non-inferiority criteria was reached. The 3-year OS (76.1% vs. 59.1%; HR of 0.69) &RFS (64.5% vs. 53%; HR of 0.71) was favouring the weekly arm as against the 3 weekly arm, confirming the non-inferiority of the lower dose weekly regimen.

#### **3.3 Definitive radiotherapy (with or without chemotherapy)**

There are no prospective trials which have directly compared primary surgery vs. primary radiotherapy in oral cavity cancers specifically. Two case series comparing these two modalities suggested a lower loco-regional control with primary radiotherapy compared to a surgical approach.

First was Studer et al., who assessed 58 consecutive oral cancer patients referred for either adjuvant or definitive radiotherapy [24]. They found local control rates were highest in the surgery followed by post-op IMRT group (92% LC at 2 yr), followed by patients treated with surgery alone or those receiving post-op 3DCRT (70–80% LC at 2 yr) and least in the definitive radiotherapy group treated with IMRT (40%LC at 2 yr) or 3DCRT (30% LC at 2 yr).

Murthy V. et al. studied 1180 oral cancer patients treated with PORT or definitive RT, by dividing them into 2 groups – Group 1 included gingiva-alveolar-buccal complex, lip and hard palate sites and group 2 included tongue and floor of mouth sites [25]. The 3 yr. LC, LRC and DFS for those treated with PORT was 74%, 65% % 60% respectively. And for those treated with definitive RT it was 34%, 31% & 30% respectively. Also, they found Group 1 patients had significantly better LC, LRC and DFS than group 2 patients.

Cohen et al. did a retrospective review of 4 multi-institutional phase II trials dealing with primary chemo-radiotherapy for T4 oral cavity tumours [26]. In all 39 patients were assessed, 42% of them having bone involvement. Median dose of RT given was 74Gy. Five-year OS, PFS and LC were 56%, 51% & 75% respectively suggesting definitive chemoradiation to be a reasonable option in these tumours.

Hosni et al. retrospectively analysed all oral cancer patients (108 patients) treated with IMRT (with no surgical intervention) [27]. The cases had 63% cT3/T4 disease, 35% cN2/3 and 35% received concurrent chemotherapy. After a median follow up of 52 months, the 5-year local, regional and distant control rate was 78%, 92% & 90% respectively. The 5-year DFS, OS and CSS were 42%, 50% & 76% respectively justifying definitive non-surgical chemoradiation as a meaningful alternative in appropriately chosen cases.

Updated MACH-NC results showed the benefit of concomitant chemotherapy was due to its effect on deaths related to head and neck cancer (absolute benefit of 9.8% in 5 years) [18]. Addition of chemotherapy showed a significant reduction in loco-regional failure (sub-HR = 0.71; p<0.0001) with a non-significant effect on distant failure.

The first Meta-Analysis of Radiotherapy in Head and neck cancer (MARCH) in 2005 and clearly showed an advantage of altered fractionation radiotherapy over conventional radiotherapy in terms of overall and progression free survival. The updated analysis of this study was conducted by Lacas et al. in 2017 by including 34 trials and 11,969 patients and extended the comparison to benefit of concomitant chemoradiotherapy over altered fractionation schemes [28]. They reiterated that altered fractionation was superior to conventional RT alone schemes in all disease indices and also that hyperfractionation schemes were better than accelerated RT plans especially when nodal disease was higher. The comparison between concomitant chemoradiation and altered fractionation revealed a clear benefit for CMT with an absolute benefit of 5.8% and 5.1% at 5 & 10 years respectively.

Combining altered fractionation RT schemes with chemotherapy in head and neck cancers have been studied in the following two trials. GORTEC 99–02 did a three-arm study comparing standard chemoradiation, accelerated radiotherapychemotherapy and very accelerated radiotherapy alone [29]. Progression free survival were 37.6%, 34.1% & 32.2% respectively with acute grade 3/4 toxicity of 76– 84% in the altered fractionation arms. The RTOG 0129 trial did a direct head on comparison between standard fractionation chemoradiotherapy and accelerated chemotherapy-radiotherapy in 743 patients [30]. At a median follow up of 7.9 years, there were no differences in OS, PFS, LRF or DM rates between the two arms. Both these studies concluded that addition of chemotherapy to altered fractionation radiation schemes in locally advanced head and neck cancers provided no benefit in terms of outcomes.
