This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Pantuck, A. J. Articles by Belldegrun, A. Search for Related Content PubMed PubMed Citation Articles by Pantuck, A. J. Articles by Belldegrun, A. Related Articles Related Reply Related Article Social Bookmarking                            What's this?

Carbonic Anhydrase IX As a Predictive Biomarker for Clear Cell Renal Cell Carcinoma

Department of Urology, University of California, Los Angeles School of Medicine, Los Angeles, CA

David Seligson

Department of Pathology and Laboratory Medicine, University of California, Los Angeles School of Medicine, Los Angeles, CA

Michael Atkins

Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA

Arie Belldegrun

Department of Urology, University of California, Los Angeles School of Medicine, Los Angeles, CA

To the Editor:

In the October 20th issue of the Journal of Clinical Oncology, Leibovich et al¹ reported on the prognostic value of carbonic anhydrase IX (CAIX) expression in clear cell renal cell carcinoma (ccRCC). In their cohort of 730 predominantly nonmetastatic ccRCC patients, low CAIX expression predicted poorer prognosis, with low levels of CAIX expression being associated with a 65% increased risk of death as a result of RCC. Additionally, CAIX expression remained a significant predictor of as a result of from RCC after adjusting individually for age, sex, presence of symptoms, TNM primary tumor classification and overall stage grouping, distant metastases, lymph node involvement, tumor size, and Eastern Cooperative Oncology Group performance status (ECOG PS). However, the statistical significance of this association was lost when simultaneous adjustment was made for TNM stage grouping, nuclear grade, and ECOG PS, leading the authors to criticize the use of CAIX as a prognostic biomarker for RCC. The authors further question the reliability of using tissue microarrays (TMAs) for CAIX analysis, having found heterogeneity in CAIX staining intensity but not in staining frequency. Although the article raises an interesting question of whether CAIX is a predictive rather than a prognostic biomarker for RCC, its findings are an outlier relative to the bulk of the related literature, and as such, the article deserves additional comment.

In contrast to the study of Leibovich et al,¹ previous studies with multivariate analyses^2,3 retained CAIX as an independent predictor of survival. In Bui et al,² analysis of CAIX was performed in a TMA constructed from 321 patients; they found that CAIX was an independent predictor after adjustment for ECOG PS, tumor stage, grade, and nodal status. In Sandlund et al,³ CAIX expression was assessed in RCC tumors using a TMA technique from 228 patients; they found that CAIX was an independent predictor when adjusted for age, sex, stage, and grade. Of note in these studies, subanalyses have shown that the survival difference may be restricted to patients with high-risk localized disease or those with metastases treated with cytokine therapy. In the article from Bui et al² that first noted this association, patients with metastatic ccRCC and less than 85% CAIX expression had a significantly worse prognosis in multivariate analysis (hazard ratio, 3.1; 95% CI, 1.99 to 4.83). Furthermore, this study suggested that patients with a complete response to interleukin-2 (IL-2)–based regimens were more likely found to have high CAIX expression (> 85%). Using an identical cut point of 85%, Atkins et al⁴ studied CAIX expression in whole tissue mount sections from 66 patients with metastatic RCC treated with IL-2, and demonstrated that tumor specimens with high CAIX expression were significantly more likely to demonstrate a complete or partial response to high-dose IL-2 (odds ratio, 3.3). This higher response rate was associated with a significant survival benefit, and survival greater than 5 years was limited to patients with high CAIX–expressing tumors. More recently, a prospective series from University of California, Los Angeles, CA,⁵ and additional series from Copenhagen⁶ and Ann Arbor,⁷ confirmed that higher CAIX expression yields a greater likelihood of response to IL-2 and a more favorable prognosis of metastatic ccRCC.⁵ Finally, in a third independent cohort from University of California, Los Angeles, CA, and Rennes, France, which examined CAIX expression in whole mount sections of 100 ccRCC patients, including 46 subjects with metastatic ccRCC treated with cytokine therapy, high CAIX expression was associated with lower T stages and absence of metastases; predicted longer progression-free survival and disease-specific survival; and was an independent prognostic factor in multivariate analysis after adjustment for ECOG PS, T stage, metastatic disease, tumor grade, and presence or absence of VHL mutation.⁸

The Leibovich series is therefore unique in finding a greater-than-expected prognostic significance of CAIX in localized RCC, and a less-than-expected significance in metastatic RCC. However, their patient composition and treatment were significantly different from those in previous reports. In the Leibovich series, for example, only 81 of the 730 patients analyzed presented with distant metastases, and only 6% of these patients received high-dose IL-2 after nephrectomy. In contrast, in the studies of Bui et al² and Atkins et al,⁴ 60% and 100% of the metastatic patients received IL-2, respectively. Moreover, in a more recent report,⁹ high CAIX was shown to predict for greater tumor shrinkage in metastatic RCC patients treated with sorafenib. Taken together, these studies raise the possibility that CAIX represents a predictive and not a prognostic biomarker. Predictive biomarkers reflect the impact of a therapeutic intervention (ie, predicts response and/or survival after chemotherapy or a biologic agent), and are therefore surrogates for response to therapy and, subsequently, for survival. Prognostic biomarkers, however, describe the natural course of a disease independent of treatment. Data from the Leibovich study raise the possibility that CAIX is less strongly associated with survival in patients with metastatic ccRCC treated with nephrectomy alone.¹ In contrast, CAIX expression may be predictive for systemic treatment response. Another issue to consider is that Bui et al² defined the cut point of 85% to distinguish low from high CAIX expression based on a cohort composed of 46% metastatic patients. The cut point has been used in subsequent studies; however, the vast majority of these studies included patients with metastatic disease treated with IL-2. Recently, this cut point has been questioned.³ Leibovich et al,¹ however, adopted the 85% cut point by convention only, and did not evaluate other possible thresholds or evaluate CAIX as a continuous variable. In the Leibovich series, low CAIX is associated with tumor aggressiveness, yet the population examined was a lower-risk cohort than was used to establish the cutoff. The results, therefore, may have been different and independent significance found had the authors used a cut point optimized to their unique patient population or had CAIX been evaluated as a continuous variable in their Cox proportional hazards model.

A final consideration relates to the use of TMA in CAIX studies. The authors of the article argue that the majority of CAIX studies were conducted on TMA and not on whole tissue sections. Although, as outlined above, this is incorrect, they hypothesize that analyses using small tissue cores in TMAs may be unreliable because of significant intratumoral staining heterogeneity. It is true that antibodies exhibiting focal and rare expression may not be best evaluated on a TMA, but CAIX expression is neither focal nor rare. Instead, ccRCC shows a uniform staining pattern.¹⁰ In Leibovich's series, 44% of tumors were purely homogeneous in staining. Moreover, the heterogeneity in staining as defined by the authors (at least 25% of a tumor section having a different staining intensity) is irrelevant for prognostic modeling.¹ The authors define heterogeneity based on staining intensity rather than expression frequency, but only the latter has been used for prognostic modeling of CAIX. As presented in their Table 1,¹ there were no tumors with weak, moderate, or marked staining that had any areas lacking staining. Hence, this so-called heterogeneity did not impact multivariate survival analysis, which was modeled based on staining frequency. Of note, we are aware of only one study that formally compared CAIX expression on matched TMA and whole sections, which found that TMA data strongly correlate with expression seen in whole sections.⁹ Finally, for many antibodies, it is possible that TMAs, because they include multiple areas averaged, may actually better capture staining heterogeneity than results obtained from a single whole tumor section. In Bui et al,² for example, calculated staining frequency was based on the mean staining from three separate TMA cores, and often tumor cores are purposely selected from more than one case block to account for tumor heterogeneity. Using a mean pooled value from cores taken across different blocks rather than using the percent positivity from a single tumor section may actually pick up and account for more rather than less staining heterogeneity.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

REFERENCES

1. Leibovich BC, Sheinin Y, Lohse CM, et al: Carbonic anhydrase IX is not an independent predictor of outcome for patients with clear cell renal cell carcinoma. J Clin Oncol 25:4757-4764, 2007[Abstract/Free Full Text]

2. Bui MH, Seligson D, Han KR, et al: Carbonic anhydrase IX is an independent predictor of survival in advanced renal clear cell carcinoma: Implications for prognosis and therapy. Clin Cancer Res 9:802-811, 2003[Abstract/Free Full Text]

3. Sandlund J, Oosterwijk E, Grankvist K, et al: Prognostic impact of carbonic anhydrase IX expression in human renal cell carcinoma. BJU Int 100:556-560, 2007[CrossRef][Medline]

4. Atkins M, Regan M, McDermott D, et al: Carbonic anhydrase IX expression predicts outcome of interleukin 2 therapy for renal cancer. Clin Cancer Res 11:3714-3721, 2005[Abstract/Free Full Text]

5. Klatte T, Zomorodian N, Kabbinavar FF, et al: Prospective evaluation of carbonic anhydrase IX (CAIX) as a molecular marker in metastatic renal cell carcinoma. J Clin Oncol 25:403s, 2007 (suppl; abstr 4601)

6. Jensen HK, Nordsmark M, Donskovl F, et al: CAIX and HIF-1a is metastatic clear cell carcinoma at baseline and during IL-2 therapy. First International CA-IX Symposium, Brussels, Belgium, November 14, 2007 (abstr)

7. Shah RB, Amin A, Braun T, et al: High Carbonic Anhydrase (CA) IX Protein Tissue Expression Predicts Response to Interleukin (IL)-2 based Therapy for Advanced Renal Cell Carcinoma Patients. J Urol 75:243 (suppl; abstr 725)

8. Patard JJ, Fergelot P, Karakiewicz PI, et al: Low CAIX expression and absence of VHL gene mutation are associated with tumor aggressiveness and poor survival of clear cell renal cell carcinoma. Int J Cancer (in press)

9. Signoretti S, Atkins M: CAIX and the treatment of metastatic clear cell carcinoma. Presented at the First International CA-IX Symposium, Brussels, Belgium, November 14, 2007

10. Potter C, Harris AL: Hypoxia inducible carbonic anhydrase IX, marker of tumour hypoxia, survival pathway and therapy target. Cell Cycle 3:164-167, 2004[Medline]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

Related Reply

• In Reply
  Bradley C. Leibovich, Yuri Sheinin, Christine M. Lohse, R. Houston Thompson, John C. Cheville, Jan Zavada, and Eugene D. Kwon
  JCO 2008 26: 3107-3109 [Full Text]

Related Article

• Carbonic Anhydrase IX Is Not an Independent Predictor of Outcome for Patients With Clear Cell Renal Cell Carcinoma
  Bradley C. Leibovich, Yuri Sheinin, Christine M. Lohse, R. Houston Thompson, John C. Cheville, Jan Zavada, and Eugene D. Kwon
  JCO 2007 25: 4757-4764 [Abstract] [Full Text]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Pantuck, A. J. Articles by Belldegrun, A. Search for Related Content PubMed PubMed Citation Articles by Pantuck, A. J. Articles by Belldegrun, A. Related Articles Related Reply Related Article Social Bookmarking                            What's this?