The Benefits of Adjuvant Trastuzumab for HER‐2‐Positive Salivary Gland Cancers

Abstract

Background

Although high‐grade salivary gland cancers (SGCs) often express androgen receptor (AR) and/or HER‐2/neu, therapeutically targeting these receptors in SGC remains investigational. We investigated the prevalence of receptor expression and the benefit of adjuvant HER‐2 directed therapy in the high‐risk postoperative setting and explored the clinical utility of sequentially targeting these receptors in the setting of advanced disease.

Materials and Methods

We clinically annotated 95 patients with SGC (excluding adenoid cystic carcinoma) treated at our institution from 2002 to 2019 and recorded AR, HER‐2/neu status, and tumor genomic profiling results when available. Clinicopathologic information was then integrated with outcomes.

Results

Of 95 patients, most had high‐risk histologies, with salivary duct carcinoma (SDC) as the most frequent diagnosis (43, 45%). Thirty‐five (37%) experienced recurrence (51% SDC). HER‐2/neu was positive (1–3+) by immunostaining in 34 of 52 (65%) evaluable cases. There was no difference in survival based on HER‐2/neu or AR expression. Nine of 17 (53%) patients with HER‐2+ SDC received adjuvant chemoradiation with trastuzumab. Median disease‐free survival (DFS) and overall survival (OS) were longer among patients with HER‐2/neu 3+ staining tumors who received adjuvant trastuzumab versus those who did not (DFS, 117 vs. 9 months; p = .02; OS, 74 vs. 43 months; p = .02), with no difference among other HER‐2/neu subgroups (0–2+). Two of nine (22%) patients treated with adjuvant trastuzumab demonstrated recurrence, both with low HER‐2/neu staining intensity (1+). Longer time to recurrence (hazard ratio, 0.94; p = .01) predicted improved outcomes. Both androgen deprivation and HER‐2‐directed therapies had clinical benefit beyond the first‐line metastatic setting, with partial response observed beyond second‐line use.

Conclusion

Although prospective data are lacking, the use of adjuvant trastuzumab in high‐risk patients with SGC appears beneficial, particularly among patients with tumors exhibiting HER‐2/neu 3+ immunostaining.

Implications for Practice

Results of this study showed an improved disease‐free and overall survival in patients treated with adjuvant trastuzumab for high‐risk salivary gland cancers with strong HER‐2/neu staining intensity. Following recurrence or metastatic spread, sequential HER‐2, and androgen‐directed therapies may benefit certain patients with salivary gland cancer.

Short abstract

This article evaluates long‐term outcomes in patients with HER2‐positive salivary gland cancer (SGC) who received adjuvant HER2‐directed therapy and in patients with recurrent or metastatic non‐adenoid cystic carcinoma SGC treated with sequential androgen deprivation therapy and/or HER2‐directed therapies over time.

Introduction

Salivary gland cancers (SGCs) represent a rare but unique group of histologically and molecularly distinct malignancies with a wide range of clinical behavior 1, 2. Despite variability in prognosis, salivary gland tumors are generally managed with a curative intent surgical approach, followed by adjuvant radiation depending on resection margins, histologic features, and nodal involvement 3, 4. The role of adjuvant concurrent chemotherapy remains unclear, with a lack of prospective data, but large retrospective analyses have failed to show an overall survival benefit 5, 6. In the recurrent, incurable, or metastatic setting there is no clear evidence that cytotoxic chemotherapy impacts the natural history of disease 7, and systemic therapy is often avoided for long periods given the slow progression noted in some SGC histologies (namely adenoid cystic carcinoma [ACC]). In an era of molecularly based cancer therapy, expression of hormone receptors and HER‐2/neu among SGCs has captured clinical interest.

With steep variation among histologies, SGCs can show variable expression of the androgen receptor (AR), although strong diffuse staining is characteristic of most cases of salivary duct carcinoma (SDC) and carcinoma ex pleomorphic adenoma 8, 9, 10. Independent of hormone expression, HER‐2/neu overexpression has been observed in up to 30% of malignant SGCs, with higher staining intensity (3+) by immunohistochemistry (IHC) often noted in SDC (with ERBB2 amplification present in only a subset of cases) 11. Borrowing from other malignancies such as prostate and breast cancer, there is strong interest and rationale for using adjuvant androgen deprivation therapy (ADT) and/or HER‐2 directed therapies, respectively, to mitigate the risk of distant metastatic spread in the high‐risk, curative setting. However, to date, limited data exist to support this approach 12, 13. In the metastatic setting, several small trials have evaluated both ADT 14, 15, 16, 17 and HER‐2‐directed therapies in SGCs 18, 19, 20, 21, with favorable response rates. The focus of the present study was to evaluate long‐term outcomes among patients with HER‐2+ SGC receiving adjuvant HER‐2 directed therapy, with mention of the use of sequential ADT and/or HER‐2‐directed therapies over time in recurrent or metastatic non‐ACC SGCs.

Materials and Methods

Patients

Following institutional review board approval for existing protocols, we reviewed electronic health record primary diagnosis codes at our institution and identified 95 patients with primary salivary gland malignancies (excluding ACC) who had even a single visit 5 years from date of query. Clinical, histopathologic, and long‐term treatment data were recorded. Immunostaining for both the AR and HER‐2/neu by IHC were documented among high‐grade histology cases. Among patients with HER‐2/neu‐positive SGC requiring adjuvant therapy, our approach is to deliver weekly carboplatin (area under the curve, 1.5) with paclitaxel (45 mg/m²) and trastuzumab (TCH; 2 mg/kg after a 6 mg/kg week 1 loading dose) intravenously concurrently with adjuvant radiotherapy (RT), then continue the adjuvant TCH for up to 12 additional weeks post‐RT, followed by a total of 1 year of maintenance trastuzumab every 21 days (6 mg/kg) 13. Twenty‐five of 95 (26%) patients had undergone targeted next‐generation massively parallel sequencing (MPS) to identify molecular targets and rearrangements.

Immunohistochemistry

HER‐2/neu antibody testing on salivary cancers (52/95, 55%) was performed using a rabbit polyclonal anti‐human antibody directed at an epitope on the cytoplasmic portion of the HER‐2 receptor (A0484, Dako, Carpinteria, CA, in the HercepTest). Immunostaining was performed on 4‐μm thick unstained slide sections and analyzed by automation (Dako Autostainer Plus). HER‐2/neu was scored by the assigned pathologist according to American Society of Clinical Oncology and College of American Pathologists breast cancer guidelines 22. In brief, scoring was as follows: negative (0), no immunoreactivity or immunoreactivity in less than 10% of tumor cells; negative (1+), faint, weak immunoreactivity in more than 10% of tumor cells but only a portion of the membrane was positive; equivocal (2+), weak to moderate complete membrane immunoreactivity in more than 10% of tumor cells; and positive (3+), moderate to strong complete membrane immunoreactivity in more than 30% of tumor cells. In some cases, fluorescence in situ hybridization was performed (Clarient, Aliso Viejo, CA) using the Pathvysion HER‐2 DNA probe kit (Abbott Laboratories, Abbott Park, IL) and reported as the average HER‐2/neu count to average chromosome enumeration probe (CEP) 17 count. The total counts were used to calculate the final ratio. A HER‐2/CEP17 ratio > 2.2 was reported as amplified. AR immunostaining (46/95, 48%) was also evaluated by IHC using the Dako AR441 antibody (1:200 dilution). A cutoff of 1% expression or greater was used to define AR positivity.

Molecular Profiling

Targeted MPS (25/95, 26%) required availability of >20% tumor on H&E slide review (6 at a site of metastatic disease, 19 from the primary tumor specimen). Tumors were sequenced on an in‐house Illumina HiSeq 2500 platform using 2 × 100 paired‐end reads evaluating the full coding regions of 447 genes and 60 selected intronic regions with importance in cancer 23, 24.

Statistical Analysis

Beyond descriptive statistics, we used Fisher's exact test (categorical variables) and Mann‐Whitney U testing (continuous variables) to compare clinical parameters by disease recurrence status. Survival was evaluated by the Kaplan‐Meier method, and log‐rank testing was applied. Disease‐free survival (DFS) was defined from the completion of curative intent therapy to date of first recurrence. Overall survival (OS) was defined from the date of initial diagnosis to death or censored at last follow‐up (OS from the date of recurrent or metastatic [R/M] diagnosis was determined when specified). Binary logistic regression was applied to identify predictors of recurrence, and the Cox proportional‐hazards model was used to investigate the association between survival and clinicopathologic parameters. All statistical tests were two sided, and a p value <.05 was considered significant. Stata 14.2 (College Station, TX) software was used.

Results

Study Population

Demographic, clinical, and pathologic characteristics of the entire SGC cohort (n = 95) are summarized in Table 1. Most had high‐risk salivary histologies (69%), with salivary duct carcinoma as the most frequent diagnosis (43/95, 45%). Nearly all patients were treated with definitive surgery with or without radiation or chemoradiotherapy (CRT) for their initial disease (83/95, 87%). Seven (7%) received systemic therapy as their initial treatment (12 patients had metastatic disease at first presentation, 9 of which were SDCs). Twenty‐eight patients (29%) received adjuvant concurrent CRT for high‐risk resected SGC, and 19 of these (68%) were SDC tested for HER‐2/neu by IHC (the remaining 9 cases were non‐SDC cases and not tested). Therefore, 19 patients (20%) received adjuvant CRT for postoperative high‐risk pathologic disease and had tumor IHC performed for HER‐2/neu (Table 2).

Table 1.

Demographics and clinicopathologic information in patients with advanced salivary gland cancers (excluding ACC)

Characteristic	All patients, n = 95	No recurrence of disease, n = 45	Recurrent or metastatic disease, n = 50	p value
Age at diagnosis (range), yr	60 (18–87)	56 (18–78)	62 (22–87)	.87
Gender, n (%)				<.01a
Female	37 (39)	25 (56)	12 (24)
Male	58 (61)	20 (44)	38 (76)
Ethnicity and race, n (%)				.92
White	86 (91)b	39 (87)	47 (94)
Asian	3 (3)	1 (2)	2 (4)
Black	4 (4)	3 (7)	1 (2)
Other or unspecified	2 (2)	2 (4)	0
Site of primary disease, n (%)				.53
Submandibular gland	16 (17)	8 (18)	8 (16)
Sublingual gland	1 (1)	1 (2)	0
Parotid gland	57 (60)	22 (49)	35 (70)
Minor salivary gland	17 (18)	12 (27)	5 (10)
Other site head and neck	4 (4)	2 (4)	2 (4)
Low risk histology (low grade), n (%)
Acinic cell	4 (4)	1 (2)	3 (6)
Adenocarcinoma	3 (3)c	3 (7)	0
Basal cell adenocarcinoma	2 (2)	0	2 (4)
Epithelial‐myoepithelial	1 (1)	0	1 (2)
Hyalinizing clear cell	2 (2)d	1 (2)	1 (2)
Mammary analogue secretory	4 (4)e	4 (9)	0
Mucoepidermoid	12 (13)f	12 (27)	0
Oncocytic carcinoma	1 (1)	0	1 (2)
High risk histology (high grade), n (%)				<.01a
Adenocarcinoma	4 (4)c	0	4 (8)
Carcinoma ex pleomorphic adenoma	10 (11)	7 (16)	3 (6)
Mucoepidermoid	3 (3)	1 (2)	2 (4)
Metastasizing pleomorphic adenoma	3 (3)	0	3 (6)
Polymorphous adenocarcinoma	3 (3)	2 (4)	1 (2)
Salivary duct carcinoma	43 (45)	14 (31)	29 (58)
Initial disease staging, n (%)				<.01a
Stage I, II	39 (41)	35 (78)	4 (8)
Stage III, IVA–C	55 (58)g	10 (22)	45 (90)
Unknown	1 (1)	0	1 (2)
Initial or primary therapy, n (%)				.88
Surgery	25 (26)	17 (38)	8 (16)
Surgery + RT	30 (32)	12 (27)	18 (36)
Surgery + CRT	28 (29)	16 (13)	12 (24)
CRT alone	5 (5)	0	5 (10)
Systemic therapy	7 (7)	0	7 (14)
Rate of recurrence	35 (42)
Patterns of recurrence, n (%)
Locoregionalh	16 (17)
Distanti	25 (26)
Lungs	13 (52)
Liver	2 (8)
Bone	11 (44)
Central nervous system	2 (8)
Androgen receptor evaluated by IHC, n (%)	46 (48)j	13 (29)	33 (66)
(+)	36 (75)	10 (77)	26 (79)
(−)	10 (25)	3 (23)	7 (21)
HER‐2/neu receptor evaluated by IHC, n (%)	52 (55)j	16 (36)	36 (72)
3+	14 (27)	6 (38)	8 (22)
2+	13 (25)	3 (19)	10 (28)
1+	7 (13)	3 (19)	4 (11)
0	18 (35)	4 (25)	14 (39)
Targeted genomic sequencing of tumor, n (%)	25 (26)	8 (18)	17 (34)
Number of lines of therapy for advanced disease,k n (%)
None, surveillance only	15 (30)
1–2	15 (30)
3–4	11 (22)
5+	9 (18)
Received locally ablative therapy	6 (6)
Received palliative radiation	21 (22)

^a p < .05, Fisher's exact testing (categorical variables), Mann‐Whitney U test (continuous variables), two‐sided.

^bFour individuals identified as Hispanic.

^cAdenocarcinoma is depicted as either high or low grade and counted once in the total cohort.

^dBoth cases with EWSR1 rearrangement confirmed by fluorescent in situ hybridization (FISH).

^eAll four cases with ETV6 rearrangement confirmed by FISH.

^fIncludes both low and intermediate grade histologies.

^gTwelve patients had metastatic disease at first presentation.

^hSix patients had both locoregional and synchronous metastatic disease.

ⁱSites of distant disease not mutually exclusive and therefore total >35.

^jAll evaluable cases demonstrated high‐grade histology.

^kAmong those individuals with recurrence or metastases (n = 50).

Abbreviations: CRT, chemoradiation; IHC, immunohistochemistry; RT, radiotherapy.

Table 2.

Adjuvant therapy in salivary duct carcinoma by HER‐2/neu status

No.	Primary site	Initial staging	Age at diagnosis	IHC	ERBB2 amplified, a (ratio)	AR	Therapy	EOT	Recurrence	DFS	OS
54	Parotid	III	78F	0	–	–	S + R + C/T	06/09/14		67.0	71.0+
69	Parotid	IVA	67F	0	–	+	S + R + C/T	02/05/15	Y	18.0	65.0+
68	Parotid	IVA	62F	1+	No	+	S + R + C/T + H	02/24/15	Y	17.0	62.0+
40	Parotid	I	60F	1+	No	Neg	S + R + C/T	08/12/16		22.0	26.0+
50	Parotid	II	63F	1+	No	+	S + R + C/T + H	12/04/19		1.0	14.0+
71	Parotid	IVA	60F	1+	–	+	S + R + cisplatin	11/06/18		13.0	18.0+
67	Submandibular	IVA	78M	1+	No	+	S + R + C/T + H	02/23/17	Y	24.0	48.0+
26	Buccal	III	72M	2+	No (1.2)	+	S + R + C/T	12/15/10	Y	4.0	101.0
41	Parotid	IVA	60M	2+	No (1.1)	+	S + R + C/T	10/21/16	Y	2.0	38.0
38	Parotid	IVA	55F	2+	No (1.6)	+	S + R + C/T + H	02/14/18		23.0	37.0+
16	Parotid	IVA	70M	2+	No (1.2)	+	S + R + C/T	04/15/16	Y	3.0	7.0+
44	Parotid	IVA	59F	3+	–	+	S + R + C/T	07/16/14	Y	0	2.0
80	Submandibular	III	83M	3+	Neg	+	S + R + C/T	06/15/08	Y	112.0	118.0
55	Parotid	III	52M	3+	–	Neg	S + R + nedaplatin	01/09/17	Y	9.0	43.0+
4	Parotid	III	61M	3+	Yes	+	S + R + C/T + H	01/15/09		123.0	137.0+
9	Parotid	IVA	62M	3+	Yes	+	S + R + C/T + H	01/15/10		117.0	131.0+
51	Parotid	IVB	69M	3+	Yes	+	S + R + C/T + H	ongoing		–	11.0+
88	Parotid	IVA	72M	3+	–	+	S + R + C/T + H	02/24/17		16.0	17.0+
95	Parotid	IVA	52M	3+	Yes	+	S + R + cisplatin + H	ongoing		–	6.0+

DFS was calculated from the date of completion of all adjuvant therapy to either first recurrence or to death/last censored follow‐up in those without recurrence. OS was calculated from initial diagnosis to death/last censored follow‐up.

^a Fluorescent in situ hybridization or next‐generation sequencing performed for ERBB2 amplification (indicate HER‐2/CEP‐17 ratio, >2.2 qualifies as amplified).

Abbreviations: –, not tested; AR, androgen receptor; C/T, carboplatin, paclitaxel; DFS, disease‐free survival in months; EOT, end of adjuvant therapy; F, female; H, trastuzumab; IHC, immunohistochemistry; M, male; OS, overall survival (in months); Neg, negative result by IHC; R, radiotherapy; S, surgery; Y, yes, indicating recurrence. “+” denotes ongoing follow‐up and the patient is alive.

Adjuvant Therapy Among HER‐2/neu‐Positive Salivary Cancers

Nine of 19 (47%) patients completed adjuvant TCH with concurrent RT, followed by up to 12 weeks of TCH and later trastuzumab alone for up to 1 year for HER‐2/neu tumor staining intensity of 1+ or greater (at the discretion of the treating physician). Median DFS and OS were both longer among TCH‐ versus TC‐treated patients with HER‐2/neu 3+ staining tumors (DFS, 117 vs. 9 months; p = .02; OS, 74 vs. 43 months; p = .04), but there was no difference among other HER‐2/neu subgroups (0–2+; Figure 1). Of interest, only two of nine (22%) patients treated with adjuvant TCH demonstrated recurrence, both with low HER‐2/neu staining intensity (1+). Moreover, none of the six HER‐2/neu 2–3+ patients treated with adjuvant TCH experienced recurrence during follow‐up. Following recurrence, there was no survival difference based on whether an individual had received adjuvant trastuzumab or not (p = .41).

Figure 1.

(A): Swimmer's plot showing outcomes among patients with salivary gland cancer (SGC) receiving adjuvant CRT with or without H. AR and/or HER‐2/neu receptor expression (staining intensity key to the left from 1–3+) by IHC arranged from shortest to longest survival or follow‐up time. Elapsed time (in months) is depicted on the x‐axis from the time of diagnosis of initial disease. (B): OS and DFS in months for patients with SGC treated with surgery followed by adjuvant CRT with or without H (n = 19) stratified based on HER‐2/neu immunostaining intensity (IHC 1 to 3+). * p < .05, multiple regression or Cox proportional hazard modeling. (C): Number of recurrences after definitive surgery and adjuvant chemoradiation (with or without H) among SGC cases (n = 9) stratified by HER‐2/neu IHC intensity. (D): OS in months from the time of first disease recurrence to death or censored at last follow‐up for patients treated with adjuvant CRT with or without H (n = 9). * p < .05, log‐rank testing.

Abbreviations: AR, androgen receptor; CRT, chemoradiation; DFS, disease‐free survival; H, trastuzumab; IHC, immunohistochemistry; OS, overall survival; R/M, recurrent or metastatic.

Recurrence and Survival Among the Entire SGC Cohort

Excluding 12 patients with metastatic disease at presentation, 35 of 83 (42%) patients in the entire SGC cohort experienced recurrence (18 [51%] had SDC), most often at lung and/or osseous distant metastatic sites. In comparing subgroups with and without disease recurrence, patients with recurrence were more often male, had high‐grade histology, and presented with more advanced initial staging on univariate analysis (all p < .01). Both high‐grade histology and nodal involvement significantly predicted recurrence (odds ratio [OR], 2.67; p = .03; OR, 1.83; p = .03, respectively) in multivariate analysis (online supplemental Table 1). Median follow‐up time was 5.2 years (range, <1–40.2), with 18 deaths observed. Male patients (p = .01), those with R/M disease (median OS, 8.4 years [95% CI 6.7–9.9], p < .01), and patients with SDC (p = .04) all had inferior survival (Fig. 2). There was no difference in survival outcomes based on initial histologic grade (p = .37), by immunostaining for AR (p = .68), or based on HER‐2/neu expression (p = .95). In multivariate analysis, female gender (hazard ratio [HR], 0.21; p = .03), uninvolved nodes (HR 2.49, p = .04), and a longer time to first disease recurrence (HR, 0.94; p = .01) were associated with improved outcomes (online supplemental Table 2).

Figure 2.

(A): OS in months among the entire salivary cancer cohort (n = 95) based on gender and (B) among those patients with R/M disease. (C): OS (in months) stratified by salivary duct carcinoma cases (n = 43) vsersus all other histologies. (D): Frequency of low and high‐grade histologic subtypes of salivary cancer among the cohort, and (E) OS by histologic grade. (F): Frequency (depicted in a 10 × 10 dot plot with AR coded as positive in blue, negative in white, and untested in shaded gray; HER‐2/neu coded in shades of purple by staining intensity [1–3+], negative in white [0], and untested in shaded gray) and OS (in months) among non‐ACC salivary cancers stratified by AR and HER‐2/neu positivity. *p < .05, log‐rank testing.

Abbreviations: AcCC, AR, androgen receptor; CA, EMC, HCCC, MASC, MEC, PA, PLGA, OS, overall survival; R/M, recurrent or metastatic; SDC,

Palliative Androgen Deprivation and HER‐2‐Directed Therapies Among Salivary Cancers

Figure 3 depicts the sequential use of palliative therapies over time among R/M patients (32/95, 34%), with a median OS of 17 months. Patients received a mean of 3.2 (range, 1–7) therapies for R/M disease. Objective response rates arranged by therapy category are depicted in supplemental online Table 3. Carboplatin with paclitaxel and cyclophosphamide, doxorubicin, cisplatin (CAP) yielded the highest response rates among cytotoxic agents with a clinical benefit rate (CBR; complete response + partial response [PR] + stable disease [SD]) ≥3 months from the time of achieving response of 60%–70%.Targeted therapies (in the context of various therapeutic protocols) demonstrated a ≥ 3 month CBR of 55%. Mutational frequencies among the sequenced patients in the cohort are also depicted in Figure 3 and often determined targeted therapy choice.

Figure 3.

(A): Swimmer's plot showing sequential therapies over time among patients with R/M salivary cancer (primary site of disease coded at left) with AR and/or HER‐2/neu receptor expression (staining intensity key to the left from 1–3+) by immunohistochemistry arranged from shortest to longest survival or follow‐up time. Elapsed time (in months) is depicted on the x‐axis from the time of diagnosis of R/M disease (the vertical dotted line represents the median survival time, 17 months). (B): OS in months from the time of first R/M disease as a function of whether or not the patient ever received ADT and/or HER‐2‐directed therapies. *p < .05, log‐rank testing. (C): Objective response rates (as a percentage) to ADT or HER‐2‐directed therapies stratified by line of therapy (from first to sixth). Number of patients per column listed above each bar. (D): Mutational plot showing commonly mutated genes among the sequenced patients (n = 25) arranged by frequency. Each column depicts a patient identifier and denotes histology. Mutational event key on right.

Abbreviations: ADT, androgen deprivation therapy; AR, androgen receptor; OS, overall survival; R/M, recurrent or metastatic.

Table 3.

Palliative HER‐2 directed therapy in HER‐2/neu(+) salivary duct carcinoma

No.	Primary site	Initial staging	Age at diagnosis	IHC	ERBB2 amplifieda	AR	Initial Therapy	DFI	Metastatic site(s)	Palliative therapies	OS
1	Parotid	IVA	53M	0	‐c	+	S + R	38.0	L	trastuzumabb	78.0
3	Hard palate	IVA	69F	1+	No (1.1)	Neg	S	255.0	D		314.0+
65	Parotid	IVC	61M	1+	No (1.3)	+	C	‐c	D	ADT	11.0
67	Submandibular	IVA	78M	1+	No	+	R + C/T	24.0	D	trastuzumab	44.0+
68	Parotid	IVA	62F	1+	No	+	S + R + C/T/H	17.0	D	trastuzumab + C/T	58.0+
18	Parotid	IVC	62F	2+	Yes (2.3)	+	S + R + letrozole	53.0	D	ADT	70.0+
26	Buccal	III	72M	2+	No (1.2)	+	S + R + C/T	4.0	D	ADT	101.0
33	Parotid	IVA	52M	2+	No (1.3)	+	S + R	110.0	D		145.0+
36	Parotid	IVA	79M	2+	No (1.6)	+	S + R	1.0	D	trastuzumab + docetaxel	7.0+
41	Parotid	IVA	60M	2+	No (1.1)	+	S + R + C/T	2.0	D		35.0+
77	Parotid	III	76M	2+	‐	Neg	S	10.0	L		16.0+
85	Submandibular	IVA	87M	2+	No (1.3)	+	R + C/T	9.0	D	ADT, trastuzumab	37.0
7	Submandibular	IVC	67M	3+	No	Neg	C	‐	D	trastuzumab + T	23.0
16	Parotid	IVA	70M	3+	‐	+	S + R + C/T	3.0	D		7.0+
21	Parotid	IVC	53M	3+	Yes	+	C	‐	D	ADT, trastuzumab, lapatinib	22.0+
29	Submandibular	IVA	60M	3+	No	‐	S + R	17.0	D		30.0+
42	Parotid	IVA	45M	3+	No	+	S + R + C/T/H	‐	D	ADT, trastuzumab	13.0
44	Parotid	IVA	59F	3+	‐	+	R + C/T	9.0	L		2.0+
55	Parotid	III	52M	3+	Yes	Neg	S + R + nedaplatin	9.0	D	trastuzumab, TDM‐1	35.0+
80	Submandibular	III	83M	3+	‐	+	S + R + C/T	112.0	L		118.0+

DFS was calculated from the date of completion of all adjuvant therapy to first recurrence. OS was calculated from initial diagnosis to death/last censored follow‐up.

^aFluorescent in situ hybridization or next‐generation sequencing performed for ERBB2 amplification (indicate HER‐2/CEP‐17 ratio, > 2.2 qualifies as amplified).

^bPatient was treated with HER‐2‐directed therapy at an outside facility despite being HER‐2/neu 0 by IHC

^cDashes (‐) denote the patient presented with metastatic disease initially.

ADT, androgen deprivation therapy; AR, androgen receptor; C, chemotherapy; C/T, carboplatin/paclitaxel; D, distant; DFI, disease‐free interval in months; EOT, end of adjuvant therapy; H, trastuzumab; IHC, immunohistochemistry; L, locoregional; Neg, negative result by IHC; OS, overall survival (in months); R, radiotherapy; S, surgery. “+” denotes ongoing follow‐up and the patient is alive.

Nine patients (28%) received HER‐2‐directed therapies following diagnosis of R/M disease, with three PRs (33%, median duration of response [DOR] 27.7 months) and five with stable disease (56%, SD) for a ≥3 month CBR of 89%; all received trastuzumab therapy and two received lapatinib and T‐DM1, ado‐trastuzumab emtansine at progression, respectively (Table 3). Additionally, 14 (44%) patients (not necessarily mutually exclusive) received ADT for their R/M disease based on tumor AR expression, with 2 (17%, median DOR 15.3 months) experiencing a PR and an additional 6 having SD (≥3 month CBR 57%) on gonadotropin‐releasing hormone agonists and androgen receptor blockers. Both ADT and HER‐2‐directed treatments had clinical benefit beyond first‐line therapy, with PRs observed even in fourth‐line use. Despite the use of targeted therapies aimed at these receptors, there was no difference in median OS from the time of disease recurrence based on whether the patient received palliative ADT or HER‐2‐directed therapy or not (p = .14).

Discussion

Adjuvant HER‐2‐Directed Therapy Appears to Benefit SGC Patients with Greater HER‐2/neu Expression Intensity

In 2013, our group reported on eight patients treated with adjuvant TCH for high‐risk SDC 13. We now report an additional nine patients in the current study who received TCH (with one patient receiving adjuvant weekly cisplatin and trastuzumab) for similar pathology. Among the current cohort, two of nine developed metastatic disease to the lungs following TCH within 24 months of completion of initial therapy, and both were HER‐2/neu 1+ (and nonamplified). None of our seven remaining TCH‐treated patients had recurred at a median follow‐up of 5.6 years (range, 1.3–12.3+), and all had HER‐2/neu 2–3+ tumor staining intensity (four were ERBB2 amplified). In our groups’ initial report, three of eight had progression or metastatic recurrence, which included the only patient with 1+ HER‐2/neu tumor staining (nonamplified). Of note, the primary HER‐2 antibody used in each series was similar. We further show that both DFS and OS were superior in patients within our adjuvant trastuzumab‐treated cohort who stained HER‐2/neu 3+ compared with those who received TC alone without HER‐2 therapy for HER‐2+ positive SDC (DFS, 117 vs. 9 months; p = .02; OS, 74 vs. 43 months; p = .02). For those who received adjuvant TCH and developed recurrence, survival beyond time of first recurrence was similar to those who received only adjuvant CRT (p = .41). Although clinical trial data are lacking, these findings establish a clearer role for adjuvant trastuzumab‐based therapy particularly in patients with HER‐2/neu 3+ SDC tumors to mitigate the risk of metastatic recurrence. Although this disease entity is overall rare, these findings raise important questions about neoadjuvant or adjuvant approaches that could combine trastuzumab with the newer recombinant HER‐2 inhibitor, pertuzumab, which blocks HER‐2 dimerization, as this has proven beneficial in breast cancer 25.

Sequential ADT and/or HER‐2‐Directed Therapies Appear Beneficial in Metastatic SGC

In evaluating the use of sequential systemic therapies over time in our recurrent, metastatic SGC cohort, objective response rates to cytotoxic agents were around 26% (anywhere from first to fourth line), and platinum‐based doublet (7/16, 44%) or triplet therapy (2/5, 40%) was most beneficial in terms of response. This is in line with published data favoring triplet CAP therapy in patients without ACC and without actionable molecular targets 26, 27, with similarly observed median DOR (6.3+ vs. 5.0 months). We were encouraged by the doublet platinum‐paclitaxel response rate in the first‐ or second‐line metastatic setting of 60% (6/10), with a median DOR of 16.1 months (range: 3.5–23.3+), particularly because paclitaxel is avoided in ACC 7. We also took note of a ≥ 3 month CBR of 55% among patients treated with targeted therapies. Our sequencing results were similar to prior reports reflecting the mutational landscape of SGC 28, with frequent PI3K/AKT/mTOR pathway alterations among SDC, for example.

We were interested in the use of sequential ADT and HER‐2‐directed therapies over time, given the availability of next‐generation therapies targeting these receptors. Recent studies have reported overall favorable response rates, particularly among HER‐2‐targeted agents when treating SGCs (online supplemental Table 4). Among nine patients with SGC receiving HER‐2 therapies in the R/M cohort, there were three PRs (33%), with a median DOR of 27.7 months and a ≥ 3 month CBR approaching 90%, which included one PR after prior ADT in the fourth‐line setting. SD was observed among two patients who received next‐line HER‐2‐directed therapy after trastuzumab (using TDM‐1 and lapatinib; median disease stability, 8.9+ months). Fourteen patients with R/M SGC received ADT, with two PRs (17%, median DOR 15.3 months) and a ≥ 3 month CBR of 57% with SD among four patients treated with newer ADT such as enzalutamide and abiraterone in second‐ and third‐line setting. Of interest, the duration of benefit to ADT in our cohort was longer among first‐line treated patients compared with later use of ADT following cytotoxic or other therapies (18.6 vs. 6.2 months, p = .02). Although speculative, it may be that androgen sensitivity is preserved earlier in the natural history of R/M SGC.

Time to Recurrence Impacts Survival in SGC

Older studies have summarized the negative clinicopathologic predictors of survival in malignant salivary cancers 29, 30, 31. Our findings were overall similar, but we did not observe an association with grade and poor outcomes as in prior studies, and we suspect this may have been due to low numbers of low‐grade tumors in our cohort (29/95, 31%). Of interest, we did not find the presence of AR or HER‐2 overexpression to be prognostic among high‐grade patients. In line with our findings, others have previously reported that HER‐2 or ERBB2 positivity does not impact DFS or OS among patients with SDC 32, 33, and the prognostic relevance of AR expression remains unclear 34, 35. We did observe that a longer time to recurrence (HR 0.94; p = .01) predicted improved outcomes, with 24 of 35 (69%) recurrences occurring within 3 years of completion of definitive therapy. Although prior work has shown that malignant SGC (namely SDC) tends to recur within 5 years of initial treatment 32, 36, we observed 26% (9/35) of recurrences after 5 years and up to decades later in some instances. This has important implications regarding long‐term surveillance in high‐risk SGC.

Although our study results are retrospective, they do provide detailed real‐world insight into long‐term outcomes of a rare subtype of head and neck cancer. We establish several important findings in the current work: (a) patients with SDC tumors showing high HER‐2/neu staining intensity are most likely to benefit from adjuvant trastuzumab as part of definitive therapy aimed at mitigating the risk of distant metastatic spread. (b) Response rates to ADT and/or HER‐2 directed therapies are favorable among R/M patients even when used beyond the first line, with clinical benefit rates exceeding 50%. These data along with results from prospective SGC trials will continue to define the importance of targeted therapy in this disease. (c) Up to a quarter of high‐grade SGCs recur late (after 5 years) after completion of initial therapy necessitating a long‐term surveillance plan.

Author Contributions

Conception/design: Glenn J. Hanna, Ji Eun Bae, Nicole G. Chau

Provision of study material or patients: Glenn J. Hanna, Jochen H. Lorch, Robert I. Haddad, Jonathan D. Schoenfeld, Danielle N. Margalit, Roy B. Tishler, Laura A. Goguen, Donald J. Annino Jr., Nicole G. Chau

Collection and/or assembly of data: Glenn J. Hanna, Ji Eun Bae

Data analysis and interpretation: Glenn J. Hanna, Ji Eun Bae, Vickie Y. Jo, Nicole G. Chau

Manuscript writing: Glenn J. Hanna, Ji Eun Bae, Jochen H. Lorch, Robert I. Haddad, Vickie Y. Jo, Jonathan D. Schoenfeld, Danielle N. Margalit, Roy B. Tishler, Laura A. Goguen, Donald J. Annino, Jr., Nicole G. Chau

Final approval of manuscript: Glenn J. Hanna, Ji Eun Bae, Jochen H. Lorch, Robert I. Haddad, Vickie Y. Jo, Jonathan D. Schoenfeld, Danielle N. Margalit, Roy B. Tishler, Laura A. Goguen, Donald J. Annino, Jr., Nicole G. Chau

Disclosures

Glenn J. Hanna: Regeneron, Sanofi, Maverick, Merck, Kura, Sanofi, Bio‐Rads, Prelude (C/A), Bristol‐Myers Squibb, Exicure, GlaxoSmithKline, NantKWest, Regeneron, Sanofi, Kartos (RF); Jochen H. Lorch: Bayer, Genentech (C/A), Bayer, Takeda, Bristol‐Myers Squibb (RF); Robert I. Haddad: Merck, Bristol‐Myers Squibb, AZ, GlaxoSmithKline, Genentech, Pfizer, Bayer (C/A), Merck, Bristol‐Myers Squibb, Kura, GlaxoSmithKline, Pfizer, Genentech (RF); Vickie Y. Jo: Merck and Co (E‐spouse works as Translational Scientist in drug development; nononcology); Jonathan D. Schoenfeld: Merck, Bristol‐Myers Squibb, Regeneron (RF), Debiopharm, Tilos, AstraZeneca, LEK, Catenion, ACI Clinical (C/A); Danielle N. Margalit: Galera Therapeutics (C/A‐relationship ended 2018); Roy B. Tishler: Regeneron (SAB), PSI/Oragenics (C/A); Nicole G. Chau: Merck, GlaxoSmithKline, Pfizer (RF), Eisai, Roche (H).

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board

Supporting information

See http://www.TheOncologist.com for supplemental material available online.

Supplemental Tables