Primary adenoid cystic carcinoma of the trachea: clinical outcome of 38 patients after interdisciplinary treatment in a single institution

Surgery was performed in 20 patients (53%). Eleven patients (55%) were diagnosed with stage 4 ACC (stage 1: n = 4 / 20%; stage 2: n = 4 / 20%; stage 3: n = 1 / 5%). All data is summarized in Table 4.

All patients with an incomplete surgical resection (R1, R2) underwent adjuvant radiotherapy as a multi-modal treatment approach (C¹²: n = 1, photons: n = 10). Furthermore, two R0-resected patients (10%) also received adjuvant radiotherapy because the tumor was close to the resection margin (C¹²: n = 0, photons: n = 2). Therefore, radiotherapy was performed in 13 (65%) of 20 patients following surgery (C¹²: n = 1, photons: n = 12).

In contrast, there was an interdisciplinary discussion about giving 18 patients (47%) definitive radiotherapy. In this group, 12 patients (67%) were diagnosed with stage 4 ACC (stage 1: n = 0 / 0%; stage 2: n = 2 / 11%; stage 3: n = 4 / 22%). Six patients underwent irradiation with C¹² (33%), whereupon five were found to have stage 4 ACC (83%). One patient (17%) had undergone surgery for ACC in 1986 and was then diagnosed with recurrent stage 2 ACC 34 years following tumor resection. Twelve patients (67%) received irradiation with photons (stage 2: n = 1 / 8%; stage 3: n = 4 / 33%; stage 4: n = 7 / 59%). A palliative setting was recommended for one patient (no treatment, 3%).

In most cases, both variants of radiotherapy included a primary plan with 3D conformal radiotherapy, IMRT, or carbon ion beam therapy as well as a boost plan with 3D conformal radiotherapy, IMRT, brachytherapy, or carbon ion beam therapy delivered to the planning target volume (PTV) in fractions from 1.8 to 5 Gy (Gy). Overall, the median dose for the PTV1 (primary plan) was 50 Gy (range: 46–63). A median dose of 16 Gy (range: 0–74) was applied to the PTV2 (boost plan). Therefore, the overall median total dose was 66 Gy (range: 48–74.4). For the irradiation with C¹² carbon ion boost, the dose to PTV1 (photons) was 50 Gy in median (range: 50–54) and the dose to the PTV2 (C¹² carbon ion) was 24 Gy in median (range: 18–24). In the photon only group, the dose to PTV1 was also 50 Gy in median (range: 46–54), and the dose to the PTV2 was 16 Gy in median (range: 8–74). Two patients received treatment with 60Gy (RBE) and 63Gy (RBE) carbon ion beam only, respectively. Two other patients received, after the primary plan (PTV1) with 50Gy photons, a brachytherapy boost with an total dose of 15Gy in 3 fractions. The resulting total dose in the C¹² group was 72 Gy in median (range: 60–74.4) and 66 Gy (range: 48–74) in the photon group. LEM 1 were used as the RBE model for the carbon ion beam therapy planning. Treatment calculation was performed with an α/β of 2Gy for organs at risk and the planning target volume. All data are summarized in Table 5.

The median follow-up time for all patients was 74.5 months. The median follow-up time for patients who underwent surgery alone was 185 months. The median follow-up time for patients who underwent surgery and adjuvant radiotherapy was 123 months. In contrast, the median follow-up times for patients after definitive C¹² treatment and photon treatment were 15.5 and 60 months, respectively. The 5-year OS of all patients was 95% (10-year: 81%). The 5-year FFLP was 96% (10-year: 83%), and the 5-year FFDP was 69% (10-year: 53%).

In patients who underwent surgery alone, the 5-year OS was 100% (10-year: 80%). The 5-year FFLP was 100% (10-year: 100%) and the 5-year FFDP was 80% (10-year: 60%). In patients who underwent radiotherapy alone, the 5-year OS was 100% (10-year: 83%). The 5-year FFLP and FFDP were 88% (10-year: 44%) and 67% (10-year: 34%), respectively. The longest follow-up period for C¹² irradiation alone was 20 months. After this period, no patient had developed local or distant progression (FFLP/FFDP: 100%). Compared to photon irradiation alone, FFLP was also 100% after 21 months (FFDP: 91%). The longest follow-up period for photon irradiation alone was 11.3 years. After this period, FFLP was 43%. FFDP was 38% after 11.3 years.

In patients who received multi-modal treatment including surgery and adjuvant radiotherapy, the 5-year OS was 84% (10-year: 84%). The 5-year FFLP was 100% (10-year: 100%), and the 5-year FFDP was 65% (10-year: 65%). One patient underwent surgery and adjuvant C¹² treatment. The follow-up period was 20 months. After this period, the patient had not developed any FFLP (FFDP). Compared to surgery and photon irradiation, FFLP was also 100% after 20 months (FFDP: 91%). The longest follow-up period for multi-modal treatment with photon irradiation was 12.3 years. After this period, FFLP was 75%. FFDP was 61% after 12.3 years.

All data is summarized in Table 6. All results for OS in the different treatment groups are summarized in Fig. 1. The FFLP and the FFDP are shown in Figs. 2 and 3, respectively.

ACCs arise most often in the salivary glands. In cases of the advanced tumor stage (T4), incomplete resection (R1, R2), or perineural invasion (Pn1), studies suggest adjuvant radiotherapy after surgery. In patients presenting with an unresectable tumor and/or severe comorbidities, definitive radiation therapy is indicated [15]. After multi-modal treatment, the 5- and 10-year OS was 94 and 91%, respectively. After definitive radiotherapy, the 5- and 10-year OS was up to 56 and 43%, respectively [16, 17].

ACC of the trachea is a very rare and slow-growing cancer that arises from the mixed seromucinous glands in the trachea. Therefore, only a few retrospective studies (particularly after radiation treatment) have been published to date. The published studies showed a good 5-year OS rate (> 70%) [18, 19]. These results are comparable with the OS in our study (5-year OS: 95%; 10-year OS: 81%). A complete R0 resection is considered the gold standard in treating tracheal ACC. Due to the infiltrative growth of ACCs into the surrounding tissue, incomplete resection margins are often observed following surgery. According to the literature, positive resection margins after surgery occur in 8–82% of all cases [6, 19, 20]. In our study, 55% (n = 11) of the operated patients had a microscopic (R1) or macroscopic residuum (R2). Although two published studies showed no significant difference in the OS of patients following completely or incompletely resected ACCs, these results must be discussed critically [1, 18]. Both published studies are based on a retrospective analysis of a few patients with ACC. In addition, the rate of adjuvant radiotherapy was different in these two studies. Adjuvant radiotherapy was performed in 75% / 100% of patients after incomplete resection. Only 33% / 0% of the patients received postoperative radiotherapy in the group of completely resected patients [1, 18]. In our study, 100% of the patients with incomplete resection (R1/2) received adjuvant radiotherapy and 22% with complete resection (R0) received adjuvant radiotherapy. Therefore, postoperative radiation after incomplete resection may be a possible explanation for the non-significant differences in OS following incomplete resection versus complete resection. In contrast, published data from the Massachusetts General Hospital in Boston showed survival benefit in patients with negative resection margins. In total, 91% of the patients had positive margins (tracheal, radial, or both), and only 9% had negative margins [19]. Furthermore, radiation treatment was confirmed in only 82% of the patients, while 17 patients with positive margins received no adjuvant radiation treatment. In our study, we found no significant prognostic factors for survival and/or LC. The small number of patients, the inhomogeneous treatment, and the different follow-up times of the subgroups could explain this. Other authors have identified tumor size, tumor location, age, surgery as an initial treatment, Pn, type of radiation, and radiation dose as possible prognostic factors for survival and/or LC [19‐21]. Most of the patients in the current study presented with locally advanced, stage 4 ACC (61%). Case reports suggest argon plasma coagulation (APC) and chemotherapy as alternative treatment options [22, 23]. However, APC is only suitable for the palliative tumor stage, and chemotherapy alone seems to be ineffective [24]. Previous studies recommend surgery as the best treatment of choice [1, 6].

However, in functionally and/or technically inoperable patients, definitive radiotherapy is a good therapeutic alternative. Some studies have shown significantly worse outcomes with only 12–17% 5-year survival rates in patients after definitive radiation treatment compared to surgery [25]. However, French data from 2018 show no significant difference between operated and non-operated patients, with 5-year survival rates of 82 and 86% [21]. In the current study, the 5- and 10-year OS rates were not significantly different between surgery followed by adjuvant radiation treatment (5-year: 92%, 10-year: 82%) and definitive radiotherapy (5-year: 100%, 10-year: 83%). Furthermore, the 5-year-FFLP was not significantly improved in patients who underwent definitive surgery (100%) compared to those who underwent radiotherapy alone (87%). But again, the results have to be considered with caution. Due to the retrospective evaluation of only a few patients and the different follow-up times of the sub-groups, there may be a bias in favor of radiotherapy. Furthermore, as patients with advanced tumor stages and a poor Karnofsky index are treated with definitive radiation, there is also a possible survival bias in favor of surgery.

Particularly in patients who present with inoperable tracheal ACC, dose escalation with C¹² is an option to improve tumor control, OS, and side effects. In addition to chordomas and chondrosarcomas, ACCs are one of the tumor entities that may benefit from escalation therapy with protons, neutrons, or C¹². Furthermore, prospective studies have shown an advantage of hadrons over photons for the treatment of ACC originating in the salivary glands [26]. Data from 58 patients with ACC of the head and neck showed significantly better LC, progression-free survival (PFS), and OS at 5 years in the group with C¹² boosts (59.6, 48.4, and 76.5%, respectively) compared to the photon only group (39.9, 27, and 58.7%, respectively) [27]. The median follow-up was 74 months in the C¹² group and 63 months in the photon group. Overall, 90% of patients in the C12 group and 94% of those in the photon group had a T4 tumor. All patients had macroscopic residual tumors at the start of treatment. There was no significant difference in OS between patients who had subtotal resection and inoperable ACC [27]. Our results confirm the advantage of C¹² irradiation compared to photon irradiation with regard to improved local (FFLP) and distant (FFDP) control. The physical and biological advantages of carbon ion beam therapy allow a dose escalation in the tumor (cumulative median dose in the current study: carbon ions: 72 Gy, photons: 66 Gy) without higher doses to the surrounding normal tissue. Therefore, the tumor control probability (TCP) is higher without increasing the normal tissue complication probability (NTCP). In addition, the higher relative biological effectiveness (RBE) yields to heavier DNA damage [28].

Overall, radiotherapy was well tolerated. The acute toxicities were mainly limited to CTCAE grades 1 and 2 with favorable outcomes. No ulceration, fistulation, or necrosis was reported after radiation treatment. Sixteen years after photon irradiation, one patient (3%) was diagnosed with a secondary malignant tumor. This might be a long-term complication after radiotherapy, as the risk for secondary malignant tumors is known to be increased [29]. However, breast cancer is the most common tumor in women. The lifetime risk for breast cancer is much higher compared to the risk for secondary malignancy after radiation treatment. Therefore, a random occurrence of breast cancer cannot be excluded. It can be speculated that the incidence of secondary malignancies might be decreased by using C¹² due to the physical advantages with lower integral doses in the normal tissue compared to radiation treatment with photons.