Adding erythropoietin (EPO) to radiation therapy (RT) in patients with head and neck squamous cell carcinoma (HNSCC) is not helpful and may be harmful, confirm long-term results of the Radiation Therapy Oncology Group (RTOG) 9903.

Anemic cancer patients fare worse than patients with normal hemoglobin levels. The tumor hypoxia therapy (and idea behind RTOG 9903) was to increase hemoglobin levels in anemic HNSCC patients, driving more oxygen to the tumor and enhancing the effects of RT.

But that didn’t pan out.
Initial results published in 2007 (Int J Rad Oncol Biol Phys. 2007;69:1008-1017) showed that adding EPO to definitive RT did not improve outcomes. This result was “unexpected and led to the study’s early closure because of a possible detrimental effect of EPO,” lead author George Shenouda, MD, associate professor, oncology and otolaryngology, McGill University Health Centre, Montréal, Quebec, Canada, said in a news release.

Long-term results, published April 1 in the International Journal of Radiation Oncology Biology and Physics, confirm that EPO is “not the appropriate treatment option” for anemic HNSCC patients, Dr Shenouda said.

The authors of an editorial agree. “We now conclude, based on the preclinical and clinical data, that EPO is not safe and should be used with caution in cancer patients, especially in the context of curative therapy,” they write.
The open-label, phase 3, randomized RTOG 9903 trial accrued 148 HNSCC patients with pretreatment hemoglobin levels between 9.0 and 13.5 g/dL and Zubrod performance status score of 0 to 2. After enrollment, four patients were considered ineligible, and three patients withdrew from the trial, leaving 141 evaluable patients.

All patients received RT alone or with standard chemotherapy and 72 received EPO (epoetin alfa, Procrit, Ortho Biotech) 7 to 10 days prior to beginning RT followed by a weekly dose of 40,000 units during treatment, unless hemoglobin levels exceeded 16 g/dL in men or 14 g/dL women.

Patients whose hemoglobin levels did not increase ≥1 g/dL after 4 doses of EPO received a dose increase to 60,000 units.

The median follow-up was 7.95 years for surviving patients and 3.33 years for all patients.

Despite an increase in hemoglobin levels from baseline with EPO, none of the differences in outcomes were statistically significant, and there was a trend toward improved outcomes without EPO. A detrimental effect “cannot be ruled out,” the authors say.

At 5 years, the estimated local-regional failure rate was 46.2% (95% confidence interval [CI], 34.5 – 58.0) with EPO vs 39.4% (95% CI, 27.7 – 51.1) without. The multivariate hazard ratio [HR] for RT+EPO vs RT was 1.40 (95% CI, 0.84 – 2.33, P = .19). Chemotherapy had no impact on the rate of local-regional failure.

Local-regional progression-free survival at 5 years was 31.5% (95% CI, 20.4 – 42.6) with EPO vs 37.6% (95% CI, 25.9 – 49.4) without (multivariate HR, 1.39; 95% CI, 0.94 – 2.05; P = .09).

Overall survival at 5 years was 36.9% (95% CI, 25.4 – 48.4) with EPO vs 38.2% (95% CI, 26.3 – 50.1) without (HR, 1.23; 95% CI, 0.83 – 1.82; P = .31). Rates of distant metastases were 15.6% (95% CI, 7.0 – 24.2) and 14.5% (95% CI, 6.1 – 22.9), respectively (HR, 1.07; 95% CI, 0.46 – 2.50; P = .88).

A Misguided Approach?

“It is important for us to be aware that EPO is a growth factor and as such, may stimulate the growth of cancer cells, resulting in decreased tumor control. Carefully designed clinical trials are required to address how to treat anemia in our cancer patients,” Dr Shenouda said.

Editorialists Todd A. Aguilera, MD, PhD, and Amato J. Giaccia, PhD, department of radiation oncology, Stanford University, California, note that tumor hypoxia has “long been understood to be an enemy of radiation therapy, and our ability to effectively target it has evaded us for many years. In hindsight, EPO was a misguided approach to target hypoxia.”

Nonetheless, they do think hypoxia “remains an attractive target,” with several caveats.

We may need to think differently about how to target hypoxia because challenges remain in profiling hypoxic tumors, measuring the flux throughout a treatment course, and effective targeting with minimal toxicity,” Drs Aguilera and Giaccia explain.

“Additionally, the diverse array of biologic activities regulated by hypoxia, some that act to reoxygenate tumors, could be counterproductive. Therefore, the future of targeting hypoxia may lie in specific hypoxia inducible factor targets such as factors of metabolic reprogramming, survival, invasion, and metastasis. Multiple targeted agents are under development and show some promise, but the next challenge will be how to combine them with radiation therapy,” Drs Aguilera and Giaccia conclude.