Anemia Predicts Poor Response to Lung Cancer Treatment

Abstract & Commentary

Synopsis: In a retrospective series of non-small cell lung cancer patients with locally advanced disease who were treated at the University of Pennsylvania with neoadjuvant chemoradiation therapy and surgery, an association of pretreatment anemia with poor response to chemoradiation was discovered. The finding raises the question of whether efforts to treat anemia early in patients receiving cancer therapy will be associated with more favorable outcomes in addition to the improved quality of life which has already been established.

Source: Robnett TJ, et al. Cancer J. 2002;8:263-267.

Anemia is common in patients with cancer and there is mounting clinical evidence that its presence is a negative prognostic factor. For example, in patients with head and neck or uterine cervical cancer, the presence of anemia has been shown to correlate with negative treatment outcomes.1,2 In this recently reported series, the success of neoadjuvant chemo-irradiation for non-small cell lung cancer was examined in the context of pretreatment hemoglobin. This was a retrospective analysis from the University of Pennsylvania in which 41 consecutive patients with clinical stage IIIA (N2 documented by mediastinoscopy or other invasive procedure) NSLC were treated preoperatively with chemotherapy (either cisplatin or paclitaxel based regimens) and radiation (median dose, 48.6 Gy). Responses were graded on a 4-point scale: 1) progressive disease before surgery (or excluded from surgery for technical reasons; 2) stable disease with resection performed, but specimen containing > 50% viable tumor; 3) partial response with specimen containing < 50% tumor; and, 4) complete response or near-complete response: R0 resection with no residual carcinoma or pT1N0 with only microscopic residual foci. Pretreatment hemoglobin levels were correlated with pathological outcomes using ANOVA and the nonparametric test for trend across ordered groups.

The mean hemoglobin level found for each of the 4 response groups noted above was: 1) 11.8 g/dL; 2) 12.1 g/dL; 3) 12.5 g/dL; and 4) 13.2 g/dL. This association was statistically significant. Thus, it was apparent that response to chemoradiation of NSLC improves with increasing hemoglobin levels.

Comment by William B. Ershler, MD

The goal of this work was to assess the relationship between anemia and pathological tumor response in a neoadjuvant setting for locally advanced NSLC and the data clearly indicate a significant association. For a number of reasons this seems like a likely conclusion, but outside of the clinical trial setting, supportive data would be unlikely available. Indeed, this retrospective analysis proved fortuitous in this regard, inasmuch as pathological specimens were available posttreatment on a relatively homogenous group of patients with a common variable being hemoglobin concentration. That stated, it is important to emphasize that this was a retrospective analysis and the sample size was relatively small. There is, without a doubt, some element of selection bias because patients included were only those who were deemed as candidates for trimodality therapy, including a likely aggressive surgical intervention. Thus, an untold number of patients with similarly advanced NSLC who were not eligible for neoadjuvant chemoradiation followed by surgery could not be included. Nonetheless, this may well be the first series to show that anemia is associated with a poorer response documented pathologically to neoadjuvant therapy for NSLC.

Like many clinical studies, the results bring more questions. In this case was the pretreatment anemia, a coincidental finding indicative of some other comorbidity that might itself be associated with reduced response? In this highly selected series, this seems unlikely. However, and obviously of great importance, would correction of the anemia prior to or during the neoadjuvant therapy improve the response rate? Some answers to this may be gleaned from a current Radiation Therapy and Oncology Group (RTOG) trial (RTOG 99-03) which randomly assigns anemic patients to receive XRT with or without recombinant human erythropoietin. Outcome measures for this study include both local control and quality of life. It would seem likely in these days of aggressive attention to supportive issues, such as quality of life and availability of effective growth factors, that such data would also eventually be available for NSLC.

Dr. Ershler of INOVA Fairfax Hospital Cancer Center, Fairfax, VA; Director, Institute for Advanced Studies in Aging, Washington, D.C.


1. Dubray B, et al. Radiology. 1996;201:553-558.

2. Warde P, et al. Int J Radiat Oncol Biol Phys. 1998; 41:347-353.