Randomized Phase III Trial of Radical Radiotherapy with Concurrent Carbogen and Nicotinamide in Locally-advanced Bladder Cancer

Reviewer: Charles B. Simone, II, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 9, 2009

Presenter: P. Hoskin
Presenter's Affiliation: Mount Vernon Cancer Centre, Northwood, United Kingdom
Type of Session: Scientific


  • Although radical cystectomy is generally considered to be the standard treatment for patients with muscle-invasive bladder cancer, radiation therapy has been shown to allow for high rates of bladder preservation with a similar overall survival as surgery.
  • A pilot study based in the United Kingdom found that a therapeutic advantage was achieved in advanced bladder cancer using accelerated radiation therapy with concurrent carbogen and nicotinamide (RT+CN) compared with radiation therapy alone. Initial work was based on in vivo experimentation demonstrating a radiosensitizing effect with these agents, allowing the radiation therapy to more easily result in tumor cell kill. Researchers found sensitization from both carbogen and nicotinamide individually when combined with radiation therapy, and that these agents appeared to work synergistically when both were administered together with radiation therapy. A recently published phase II trial demonstrated excellent local control of muscle invasive bladder transition cell carcinoma without a significant increase in acute or late treatment toxicity with this treatment combination [Hoskin P, et al. Int J Radiat Oncol Biol Phys. 2009;73(5):1425-31.].
  • Carbogen is a mixture of carbon dioxide and oxygen gas. When carbogen is inhaled, the increased level of carbon dioxide causes a physiological response of the body to generate more oxygen by increasing the respiratory rate and heart rate and by causing deep breathing. On a cellular level, alkaline buffering agents are released to remove carbonic acid from the bloodstream.
  • Nicotinamide is a water-soluble vitamin that is part of the vitamin B group that can act as a chemo- and radiosensitizing agent by enhancing tumor blood flow, thus reducing tumor hypoxia.  Niacinamide can also inhibit poly(ADP-ribose) polymerase (PARP-1) enzymes involved in the rejoining of DNA strand breaks that are induced by radiation therapy or chemotherapy.
  • Based on these in vivo and phase II findings, between 2000 and 2006, 333 patients in the United Kingdom were recruited into a Phase-III randomized trial comparing RT+CN (n = 168) and radiotherapy alone (n = 165).

Materials and Methods

  • Eligibility criteria included patients over the age of 18 years able to comply with carbogen breathing with transitional cell carcinoma of the bladder with either muscle-invasive carcinoma Stage T2 or T3 of any grade (87%), high-grade superficial carcinoma T1 (10%), or prostatic invasion T4a (3%).
  • Following a maximum transuretheral resection of the bladder tumor, radiation therapy was administered to patients to doses of 55 Gy in 20 fractions over four weeks or to 64 Gy in 32 fractions over six-and-a-half weeks using a three-field or four-field technique.
  • Carbogen (2-5% CO2 + 95-98% O2) breathing for patients randomized to the RT + CN arm was conducted via a closed breathing system, with carbogen administered at 15 breaths per minute during the five minutes prior to and during radiation therapy.
  • Nicotinamide (Larkhall Laboratories, UK) was administered orally at 40 mg/kg to 60 mg/kg at 90 minutes to 120 minutes prior to each fraction of radiation therapy.
  • The study primary endpoints were overall survival, disease-specific survival, and local relapse-free survival. Local response and relapse-free survival were assessed via cystoscopy. Secondary endpoints assessed bowel and rectal morbidity, scored using the Common Toxicity Criteria scales.



  • Patients had T2 tumors more often than T3 tumors, and few patients had T1 lesions. Most patients had Grade 3 tumors. 
  • Overall compliance to radiation therapy was 98%.  Among patients randomized to the RT+CN arm, compliance with nicotinamide was 70% and compliance with carbogen administration was 85%.
  • Three-year overall survival was higher for patients randomized to RT+CN when compared to those in the radiation therapy alone arm (59% versus 46%, p = 0.04). The median survival was 54 months in the RT+CN arm and 30 months in the radiation therapy alone arm. Disease-specific survival trended higher for patients in the RT+CN arm (68% versus 56%, p = 0.11).  The hazard ratio was 0.8613 (CI, 0.7445–0.9955; p = 0.04) for risk of death from any cause and 0.8683 (CI, 0.7307–1.0304; p = 0.1) for risk of death from local or metastatic disease, both in favor of RT+CN.  Local relapse-free survival also trended high for patients in the RT+CN arm (74% versus 63%, p = 0.11). No significant difference was seen in any primary endpoint among patients treated with radiation therapy to 55 Gy when compared to those treated to 64 Gy.
  • There was no difference between to two arms in the frequency of Grades 1 to 4 bowel toxicity for endpoints of tenesmus, mucosal loss, sphincter control, and pain. Patients in the RT+CN arm, however, did have a higher incidence of Grade ≥2 stool frequency (10% versus 5%, p = 0.05). No difference was observed in the frequency of Grades 1 to 4 urinary toxicity for endpoints of dysuria, urinary frequency, hematuria, incontinence, and decreased stream. No significant difference was seen in any bowel or bladder toxicity endpoints among patients treated to 55 Gy when compared to those treated to 64 Gy.

Author's Conclusions

  • A 13% survival advantage at 3 years was seen in patients treated with RT+CN when compared with patients treated with radiation therapy alone.
  • Disease-free survival and relapse-free survival were higher in the RT+CN arm but did not reach significance, possibly due to the study being underpowered.
  • Morbidity was similar in both arms, indicating a therapeutic benefit with this approach.
  • The combination of carbogen and nicotinamide as radiosensitizing agents has been confirmed and should be employed in further trials in disease sites other than bladder cancer.

Clinical/Scientific Implications

  • This study further confirms that radiation therapy is a viable alternative to radical cystectomy in appropriately selected patients with muscle invasive bladder cancer.
  • Three means to improve tumor control were employed in this study beyond standard fractionation radiation therapy. Accelerated radiation therapy was used in an attempt to overcome tumor cell proliferation, carbogen was used to overcome diffusion-limited hypoxia, and nicotinamide was used to minimize capillary bed shutdown and reduce perfusion-related acute hypoxia.
  • This study of carbogen and nicotinamide was appropriately conducted in patients with transitional cell bladder carcinoma since previously reported immunohistochemical studies using markers of cellular hypoxia have confirmed a median hypoxic fraction of 10% in a range of transitional cell bladder carcinomas, and patients with higher levels of these markers were found to have inferior survival. 
  • This study demonstrated an overall survival benefit without a significant benefit in local control for patients receiving carbogen and nicotinamide. As no chemotherapy was administered, the lack of benefit in local control in patients treated with localized radiation therapy is unclear, particularly in light of a demonstrated survival benefit. It may be that the study was underpowered to determine a small benefit in local relapse-free survival. However, this may also be explained based on previous reports that hypoxia predicts both local control and tumor aggressiveness and metastatic potential. 
  • The authors are currently analyzing immunohistochemical studies and hypoxic gene probes of >300 patient samples collected during this trial to determine if tumors with increased hypoxia exhibit a greater response to carbogen and nicotinamide.
  • Other than a transient increase in stool frequency in a minority of patients in the experimental arm, no significant increase in toxicity was identified in this trial. While carbogen administration is labor intensive to administer, and not all patients will be able to tolerate this breathing, this trial indicates that carbogen can be administered safely and effectively with a compliance rate of 85%
  • The United Kingdom may proceed with further studies in additional disease sites assessing the combination of radiation therapy with carbogen and nicotinamide. Attention should be paid to determine if there will be increased rates of normal tissues toxicity in patients treated with this combination for primary disease sites other than bladder cancer.
  • Ultimately, while these results are very promising, this treatment regimen of radiation therapy, carbogen, and nicotinamide is unlikely to be widely employed in the United States in the near future. The modern nonsurgical treatment of patients with bladder cancer consists of chemotherapy delivered either neoadjuvantly35 , Advanced Bladder Cancer Overview Collaboration. Neoadjuvant chemotherapy for invasive bladder cancer, Cochrane Database Syst Rev (2005) CD005246. or concurrently with radiation therapy. As neither carbogen nor nicotinomide have cytotoxic activity, and they instead appear to function completely through increasing radiation sensitization, many practitioners in the Unites States would be very hesitant to treat patients without a systemic cytotoxic therapy. However, as hypoxia has previously been shown to impair both radiation therapy and chemotherapy responses, trials should be initiated examining the combination of carbogen and nicotinamide with both radiation therapy and chemotherapy for patients with transitional cell carcinoma of the bladder, with careful attention paid to normal tissues toxicity.