Prognostic value of CAIX expression in oral squamous cell carcinoma: a systematic review and meta-analysis

Figures & data

Figure 1. Flowchart for inclusion of the studies according to the PRISMA guidelines.

Figure 2. Forest plot for the association of higher CAIX expression with overall survival (A), disease‐free survival (B).

Table 1. Characteristics of the included studies.

Study	Year	Country	Sample size	Tumour subsite	Recruitment period	Treatment	Follow-up (months)	CAIX antibody	IHC pattern	Cut-off point (%)	CAIX (+)
Sakata K et al. [27]	2008	Japan	68	Tongue	1987–2004	Rx	11–146 (mean 56)	Rabbit polyclonal antibody to CAIX (Novus Biologicals)	Membrane	10	32.35% (n = 22)
Kim SJ et al. [28]	2007	South Korea	60	Tongue	1997–2004	Sx, and Rx	4.1–117.13 (mean 29.51)	Anti-CAIX mouse monoclonal antibody clone M75	Cytoplasmic membrane	10	63.33% (n = 38)
Choi SW et al. [29]	2008	South Korea	117	Buccal mucosa, gingiva, tongue, retromolar trigone, palate, and lip.	1996–2000	Sx, and Rx	2–120 (mean 39.5)	Anti-CAIX mouse monoclonal antibody clone M75	Cytoplasmic membrane	5	58.12% (n = 68)
Roh JL et al. [30]	2009	South Korea	21	Tongue	1997–2005	Sx, and Rx	37–123 (mean 60)	Monoclonal antibody CAIX (AF 2188)	Cytoplasmic membrane	10	60.47% (n = 26)
Eckert AW et al. [31]	2010	Germany	80	Hard and soft palate, buccal mucosa, tongue, floor of the mouth, mandibular angle, and lower alveolar mucosa	1994–1999	Sx	60	Monoclonal antibody CAIX (HI-20, SC25599)	Cytoplasmic membrane	10	42.5% (n = 34)
Han MW et al. [32]	2011	South Korea	33	Tongue	2001–2006	Sx, and Rx	40 (range 9–113)	Monoclonal antibody CAIX (AF 2188)	Cytoplasmic membrane	10	63.6% (n = 21)
Kondo Y et al. [33]	2011	Japan	107	Tongue, gingiva, floor of mouth, lips, and buccal mucosa	1992–2009	Sx	60	Rabbit polyclonal to CAIX (Abcam 15086)	Membrane	10	91.58% (n = 98)
Brockton NT et al. [34]	2012	Canada	61	Unspecified oral cancer	1998–2005	Sx and Rx	60	Rabbit polyclonal to CAIX (Abcam 15086)	N/A	*	26.23% (n = 16)
Heo K. [35]	2012	South Korea	62	Tongue and others	2003–2006	N/A	52.2 (range 5.75–86.9)	Anti-CAIX mouse monoclonal antibody clone M75	Cytoplasmic membrane	10	69.35% (n = 41)
Pérez-Sayáns M et al. [36]	2012	Spain	50	Buccal mucosa, soft palate, gums, retromolar trigone, tongue, and floor of the mouth	2006–2010	Rx, Qx, and Sx	28.7–37.9	Anti-CAIX mouse monoclonal antibody clone M75	Cytoplasmic membrane	10	82.00% (n = 41)
Hwa JS et al. [37]	2015	South Korea	24	Tongue	1998–2009	Rx, Qx, and Sx	11–116 (mean 56)	Rabbit polyclonal antibody to CAIX (Novus Biologicals)	Membrane	10	20.83% (n = 5)
Yang JS et al. [38]	2015	Taiwan	271	Unspecified oral cancer	2000–2006	N/A	150 months	Anti-CAIX antibody (Santa Cruz Biotechnology)	N/A	N/A	41.69% (n = 113)
Vasconcelos MG et al. [39]	2015	Brazil	57	Tongue	1995–2007	N/A	N/A	Anti-CAIX antibody (Santa Cruz Biotechnology)	Cytoplasmic membrane	10	66.7% (n = 38)
Simoes-Sousa S et al. [40]	2016	Brazil and Spain	135	Tongue, floor of mouth, buccal mucosa, gingiva, and retromolar trigone	NA	Rx, Qx, and Sx	105 months	Antibody CAIX (Abcam 15086)	Cytoplasmic membrane	5	57.78% (n = 78)
Brockton NT et al. [41]	2017	USA	168	Tongue, floor of mouth, buccal mucosa, and gingiva	2003–2012	Rx, Qx, and Sx	33 (range 0.2–111.0)	Rabbit polyclonal CAIX (Abcam 15086)	N/A	*	25.00% (n = 42)
Sáenz-de-Santa-María I et al. [42]	2017	Spain	108	Tongue	N/A	N/A	N/A	Antibody Carbonic CAIX (Abcam 15086)	Cytoplasmic membrane	23	48.15% (n = 52)
Peterle GT et al. [43]	2018	Brazil	52	Unspecified oral cancer	2002–2008	Sx and Rx	24–60 months	Antibody CAIX (Abcam 108351)	Cytoplasmic membrane	25	50.00% (n = 26)
Eckert AW et al. [44]	2019	Germany	158	Unspecified oral cancer	1997–2015	Sx	105 months	Anti-CAIX mouse monoclonal antibody clone M75	N/A	51	15.19% (n = 24)

N/A: not available; Qx: chemotherapy; Rx: radiotherapy; Sx: surgery

*Raw AQUA score distributions for CAIX were evaluated: high CAIX expression was defined as an AQUA score within the upper quartile. Low CAIX expression was defined as an AQUA score within the lower three quartiles.

Table 2. Synthesis of data extracted from the included studies related to outcomes pooled in the meta‐analysis.

Study	OS (HR 95% CI)	DFS (HR 95% CI)	LC (HR 95% CI)	DSS (HR 95% CI)	Adjustment
Sakata K et al. [27]	NR	NR	0.91 (0.32–2.61)	NR	Multivariate adjusted for T stage and microvessel density.
Kim SJ et al. [28]	2.99 (1.39–6.45)	1.76 (0.89–3.51)	NR	NR	None
Choi SW et al. [29]	1.91 (0.77–4.71)	1.77 (0.56–5.56)	NR	NR	None
Roh JL et al. [30]	NR	NR	1.09 (0.43–2.76)	0.71 (0.23–2.22)	None
Eckert et al. [31]	1.34 (0.65–2.76)	NR	NR	NR	Multivariate adjusted for tumour size and tumour grade.
Han MW et al. [32]	0.65 (0.12–3.67)	0.80 (0.50–3.15)	NR	NR	Multivariate adjusted for tumour size and tumour grade.
Kondo Y et al. [33]	3.36 (0.97–11.70)	NR	NR	NR	None
Brockton NT et al. [34]	NR	NR	NR	2.96 (1.01–8.66)	Multivariate adjusted for tumour stage and nodal involvement.
Heo K et al. [35]	NR	6.82 (1.22–37.94)	NR	NR	None
Pérez-Sayáns M et al. [36]	1.36 (0.43–4.26)	NR	NR	2.04 (0.76–5.49)	None
Hwa JS et al. [37]	NR	NR	NR	0.29 (0.05–1.77)	None
Yang JS et al. [38]	1.76 (1.07–2.87)	NR	NR	NR	None
Vasconcelos MG et al. [39]	0.86 (0.23–3.26)	NR	NR	NR	None
Simoes-Sousa S et al. [40]	0.76 (0.44–1.38)	NR	NR	NR	None
Brockton NT et al. [41]	NR	NR	NR	1.0 (1.00–1.01)	Multivariate adjusted for tumour stage and nodal involvement.
Sáenz-de-Santa-María I et al. [42]	1.14 (0.69–1.89)	NR	NR	NR	Multivariate adjusted for tumour stage and nodal involvement.
Peterle GT et al. [43]	NR	8.75 (0.99–77.19)	NR	2.84 (1.02–7.87)	Multivariate adjusted for tumour size and the use of Rx.
Eckert AW et al. [44]	1.70 (0.97–2.85)	NR	NR	NR	Multivariate adjusted for tumour stage, lymph node status, and tumour grade.

NR: no report.

Table 3. Quality score according to the REMARK guidelines.

Study	Year	Samples	Clinical data of the cohort	Immunohisto chemistry	Prognosis	Statistics	Classical prognostic factors	Overall risk of bias
Sakata K et al. [27]	2008	A	I	A	I	I	I	2
Kim SJ et al. [28]	2007	A	A	A	A	A	A	6
Choi SW et al. [29]	2008	A	A	A	I	A	A	5
Roh JL et al. [30]	2009	A	A	A	A	A	A	6
Eckert AW et al. [31]	2010	A	A	A	N/A	I	I	3
Han MW et al. [32]	2011	A	I	A	I	A	A	4
Kondo Y et al. [33]	2011	A	A	A	A	A	A	6
Brockton NT et al. [34]	2012	A	I	A	A	A	A	5
Heo K et al. [35]	2012	I	A	A	I	A	A	4
Pérez-Sayáns M et al. [36]	2012	A	A	A	I	A	A	5
Hwa JS et al. [37]	2015	I	A	A	I	I	I	2
Yang JS et al. [38]	2015	I	A	A	I	A	A	4
Vasconcelos MG et al. [39]	2015	A	I	A	I	I	A	3
Simoes-Sousa S et al. [40]	2016	I	A	A	A	A	A	5
Brockton NT et al. [41]	2017	A	A	A	A	A	A	6
Sáenz-de-Santa-María I et al. [42]	2017	A	I	A	A	A	I	4
Peterle GT et al. [43]	2018	A	A	A	I	A	A	5
Eckert AW et al. [44]	2019	A	A	A	A	A	A	6

Items were assessed as A: Adequate; I: Inadequate; N/A: no description.

Table 4. Pooled hazard ratios and 95% confidence intervals.

	Number of studies	Pooled HR (95% CI), fixed effects	Pooled HR (95% CI), random effects	Ri^*	Q test p Value
Overall survival
Overall	11	1.45 (1.17–1.80)	1.46 (1.12–1.89)	0.26	.21
High quality	6	1.48 (1.18–1.86)	1.51 (1.11–2.06)	0.38	.13
Low quality	5	1.21 (0.64–2.29)	1.21 (0.64–2.29)	0.00	.57
Full adjustment	4	1.48 (1.18–1.86)	1.51 (1.11–2.06)	0.38	.13
Asian	5	2.01 (1.42–2.86)	2.01 ( 1.42–2.86)	0.00	.45
Non-Asian	6	1.19 (0.91–1.57)	1.19 (0.91–1.57)	0.00	.45
Tongue	4	1.18 (0.84–1.66)	1.22 (0.66–2.27)	0.66	.04
Mixed subsites	7	1.48 (1.14–1.91)	1.48 (1.10–2.00)	0.22	.29
Use of M75 antibody	3	1.93 (1.29–2.88)	1.93 (1.29–2.88)	0.00	.39
Use of other antibodies	8	1.29 (1.00–1.67)	1.29 (0.96–1.75)	0.21	.27
Use of 10% cut-off point	6	1.72 (1.14–2.59)	1.69 (1.06–2.69)	0.17	.31
Use of other cut-off points	5	1.36 (1.06–1.75)	1.36 (0.98–1.88)	0.39	.17
Disease-free survival
Overall	5	1.98 (1.18–3.32)	2.12 (1.02–4.43)	0.40	.19
Asian	4	1.81 (1.06–3.09)	1.82 (0.90–3.69)	0.34	.25
Tongue	2	1.51 (0.79–2.88)	1.27 (0.43–3.74)	0.57	.21
Mixed subsites	3	3.24 (1.35–7.76)	3.58 (1.26–10.12)	0.26	.39
Use of 10% cut-off point	3	1.82 (0.99–3.33)	1.85 (0.59–5.77)	0.65	.12
Use of other cut-off points	2	2.50 (0.91–6.89)	2.98 (0.69–12.91)	0.48	.20
Locoregional control
Overall	2	1.01 (0.50–2.02)	1.01 (0.50–2.02)	0.00	.80
Disease specific survival
Overall	6	1.00 (0.99–1.01)	1.35 (0.78–2.33)	1.00	.03
Asian	2	0.55 (0.21–1.44)	0.55 (0.21–1.44)	0.00	.41
Non-Asian	4	1.00 (0.99–1.01)	1.79 ( 0.90–3.54)	1.00	.01
Tongue	2	0.55 (0.21–1.44)	0.55 (0.21–1.44)	0.00	.41
Mixed subsites	4	1.00 (0.99–1.01)	1.79 (0.90–3.54)	1.00	.02

*R_i stands for the proportion of the total variance due to between studies variance.

Figure 3. Funnel plot of publication bias for higher CAIX expression and overall survival.

Sakata K, Someya M, Nagakura H, et al. Brachytherapy for oral tongue cancer: an analysis of treatment results with various biological markers. Jpn J Clin Oncol 2008;38:402–7.

 PubMed Web of Science ®Google Scholar

Kim SJ, Shin HJ, Jung KY, Baek SK, et al. Prognostic value of carbonic anhydrase IX and Ki-67 expression in squamous cell carcinoma of the tongue. Jpn J Clin Oncol 2007;37:812–9.

 PubMed Web of Science ®Google Scholar

Choi SW, Kim JY, Park JY, et al. Expression of carbonic anhydrase IX is associated with postoperative recurrence and poor prognosis in surgically treated oral squamous cell carcinoma. Hum Pathol 2008;39:1317–22.

 PubMed Web of Science ®Google Scholar

Roh JL, Cho KJ, Kwon GY, et al. The prognostic value of hypoxia markers in T2-staged oral tongue cancer. Oral Oncol 2009;45:63–8.

 PubMed Web of Science ®Google Scholar

Eckert AW, Lautner MH, Schütze A, et al. Co-expression of Hif1alpha and CAIX is associated with poor prognosis in oral squamous cell carcinoma patients. J Oral Pathol Med 2010;39:313–7.

 PubMed Web of Science ®Google Scholar

Han MW, Lee HJ, Cho KJ, et al. Role of FDG-PET as a biological marker for predicting the hypoxic status of tongue cancer. Head Neck 2012;34:1395–402.

 PubMed Web of Science ®Google Scholar

Kondo Y, Yoshikawa K, Omura Y, et al. Clinicopathological significance of carbonic anhydrase 9, glucose transporter-1, Ki-67 and p53 expression in oral squamous cell carcinoma. Oncol Rep 2011;25:1227–33.

 PubMed Web of Science ®Google Scholar

Brockton NT, Klimowicz AC, Bose P, et al. High stromal carbonic anhydrase IX expression is associated with nodal metastasis and decreased survival in patients with surgically-treated oral cavity squamous cell carcinoma. Oral Oncol 2012;48:615–22.

 PubMed Web of Science ®Google Scholar

Heo K, Kim YH, Sung HJ, et al. Hypoxia-induced up-regulation of apelin is associated with a poor prognosis in oral squamous cell carcinoma patients. Oral Oncol 2012;48:500–6.

 PubMed Web of Science ®Google Scholar

Pérez-Sayáns M, Suárez-Peñaranda JM, Pilar GD, et al. Expression of CA-IX is associated with advanced stage tumors and poor survival in oral squamous cell carcinoma patients. J Oral Pathol Med 2012;41:667–74.

 PubMed Web of Science ®Google Scholar

Hwa JS, Kwon OJ, Park JJ, Woo SH, et al. The prognostic value of immunohistochemical markers for oral tongue squamous cell carcinoma. Eur Arch Otorhinolaryngol 2015;272:2953–9.

 PubMed Web of Science ®Google Scholar

Yang JS, Lin CW, Chuang CY, et al. Carbonic anhydrase IX overexpression regulates the migration and progression in oral squamous cell carcinoma. Tumour Biol 2015;36:9517–24.

 PubMed Web of Science ®Google Scholar

Vasconcelos MG, Vasconcelos RG, Pereira de Oliveira DH, et al. Distribution of hypoxia-inducible factor-1α and glucose transporter-1 in human tongue cancers. J Oral Maxillofac Surg 2015;73:1753–60.

 PubMed Web of Science ®Google Scholar

Simões-Sousa S, Granja S, Pinheiro C, et al. Prognostic significance of monocarboxylate transporter expression in oral cavity tumors. Cell Cycle 2016;15:1865–73.

 PubMed Web of Science ®Google Scholar

Brockton NT, Lohavanichbutr P, Enwere EK, et al. Impact of tumoral carbonic anhydrase IX and Ki-67 expression on survival in oral squamous cell carcinoma patients. Oncol Lett 2017;14:5434–42.

 PubMed Web of Science ®Google Scholar

Sáenz-de-Santa-María I, Bernardo-Castiñeira C, Secades P, et al. Clinically relevant HIF-1α-dependent metabolic reprogramming in oropharyngeal squamous cell carcinomas includes coordinated activation of CAIX and the miR-210/ISCU signaling axis, but not MCT1 and MCT4 upregulation. Oncotarget 2017;8:13730–46.

 PubMedGoogle Scholar

Peterle GT, Maia LL, Trivilin LO, et al. PAI-1, CAIX, and VEGFA expressions as prognosis markers in oral squamous cell carcinoma. J Oral Pathol Med 2018;47:566–74.

 PubMedGoogle Scholar

Eckert AW, Horter S, Bethmann D, et al. Investigation of the prognostic role of carbonic anhydrase 9 (CAIX) of the cellular mRNA/protein level or soluble CAIX protein in patients with oral squamous cell carcinoma. Int J Mol Sci 2019;20:375.

 Web of Science ®Google Scholar