H. Lee Moffitt Cancer Center & Research Institute

Combinations of Radiotherapy and Chemotherapy: Strange Bedfellows or a Marriage Made in Heaven?

Much of the literature of alchemy, whether Christian, Islamic, or Taoist, centers on the conjunction of opposites. Such narratives abound with references to the "marriage" or "union" of opposing principles, and the accompanying illustrations leave little doubt that this was to be a physical as well as a philosophical intermingling.

A similar metaphoric intermingling exists in the current oncologic literature. Radiation and chemotherapy, previously dour antagonists in the North American oncologic community, have found themselves to be close if not always wise or compatible bedfellows. We currently see in the oncologic literature a proliferation of terms - chemoradiotherapy, radiochemotherapy, induction chemotherapy, concurrent chemoradiotherapy, anterior chemotherapy, etc - that clearly indicates a desire for some combination but little clarity as to how it might best be accomplished. On the downside, I recall a distinguished surgical oncologist during my training in Boston who referred to all of this preoperative "stuff" as pretreatment chemotherapy.

There is a long and checkered history of attempts to combine radiation and chemotherapy with the hope of producing specific interactions that can enhance cytotoxicity in the tumor without doing so in the critical normal tissues. In 1979, Steel published a key paper in which he proposed a nomenclature for mechanisms by which radiation and chemotherapy might be combined.[1]

Spatial cooperation refers to the situation in which disease in an anatomic site not treated adequately with one modality is treated with the other. Spatial cooperation could be applied either to "pharmacologic sanctuaries" (eg, prophylactic cranial irradiation in pediatric acute lymphoblastic leukemia) or to sites of initial bulk disease (eg, consolidation radiation in small cell lung cancer or intermediate-grade lymphomas following chemotherapy).

Toxicity independence is the use of multiple therapeutic agents whose antitumor effectiveness is additive but whose toxicities to normal tissues are at least partially independent. This principle underlies the use of multidrug chemotherapy as well as combinations of radiotherapy and chemotherapy. Specific interaction among agents is not required. Although originally described by Steel in terms of a uniform population of tumor cells, toxicity independence may be expanded to consider the case in which the tumor contains subpopulations that are resistant to one modality but not to both.

Protection of normal tissues such as bone marrow has been demonstrated in the laboratory by administering one chemotherapeutic agent prior to a second, thereby reducing the net toxicity that occurs if only the second agent is given. In theory, this might work with radiation as well, but clinical applications have not been developed with multiple drugs or with drug and radiation.

Enhancement of tumor response refers to combining a drug without intrinsic cytotoxicity with radiation to obtain greater cell kill than with radiation alone. In combining two cytotoxic agents, the determination of whether the resultant cell kill is additive or supra-additive becomes more complex, particularly with nonlinear dose-survival curves. Much of Steel's paper is devoted to clarification of this issue. The general use of the term "radiosensitization" by an agent with its own cytotoxicity can be misleading and has implications regarding the need for concurrent treatment that may not be supported by available data.

Spatial cooperation, which simply requires that radiation kill tumor cells in one part of the body and systemic chemotherapy kill them in others, is responsible for most of our clinical gains to date. The classic example of spatial cooperation is the use of prophylactic cranial irradiation in pediatric acute lymphoblastic leukemia, in which radiation is used to treat a relative drug sanctuary site. Even this example may be more complicated than originally thought, as there is some evidence that cranial irradiation may reduce systemic relapse as well, at least in certain high-risk patients.[2] While there have been claims of supra-additive toxicity (or synergy) with various radiation-drug or drug-drug combinations, more careful analysis of these data by isobologram most often has found them to be additive.

Even when it is possible to produce in the laboratory a specific mechanism of enhancement, such as the kinetic optimization of radiation by synchronizing cells with an agent interacting with the mitotic spindle (vinca alkaloids 20 years ago, the taxanes now), translation into a clinical gain may be difficult. Synchronization therapy did not work well 20 years ago with the vinca alkaloids, and there is no reason to expect it to work well now. Tumor kinetic heterogeneity, as well as sensitization of normal tissue in addition to tumor, were problems then and remain problems now.[3,4] The fact that acute esophageal toxicity is dose-limiting with the concurrent use of chest radiation and paclitaxel strongly suggests that the kinetic enhancement of radiation effects is far from tumor-specific.[5]

We are at the threshold of understanding a new set of mechanisms and common pathways for radiation and drug cytotoxicity, as well as possibly enhancing them by tumor-specific genetic changes. Targeting antigenically distinct oncogene products, such as mutant erb B or ras, shows promise in the laboratory and exploits epitopes not seen in normal tissues.[6,7] The use of inhibitors of ras farnesylation not only is selectively cytotoxic for tumors expressing mutant ras proteins, but also sensitizes these cell lines to ionizing radiation.[8] The use of methylxanthines to abrogate the G₂-M checkpoint in cell lines that have lost normal G₁-S checkpoint function by virtue of p53 mutation is another example of a therapeutic strategy that targets tumor-specific genetic changes.

The present enthusiasm for concurrent chemoradiotherapy should be tempered with recognition of the known enhancement of acute toxicities and the less well-characterized potential for worsened late effects, including second malignancies, compared with sequential therapy. There is a need for well-designed clinical trials that not only focus on issues of sequencing of modalities, but also quantitate the gains and toxicities of concurrent treatment approaches. While relatively few such studies have been conducted in the past, increasing recognition of this need has led to several current trials in lung cancer that focus on exploring these issues. Addressing the following points will promote greater clarity to the science of chemoradiation:

1. Delineation of clinical strategies, intended mechanisms and outcomes, and appropriate endpoints.
2. Clarification of terminology. Administering radiation and chemotherapy on the same day with a resulting increase in toxicity does not mean that the drug is acting as a radiosensitizer. Is there any evidence (laboratory or clinical) that these effects are time-dependent and require concurrent rather than sequential therapy? The issue of differential sensitization of tumor and normal tissues also needs to be studied.
3. A better understanding of common molecular pathways mediating radiation and drug-initiated cell damage, damage repair, and death. Since nature rarely operates with a single mechanism, we must be cautious in equating enhancement of a means of cell death (eg, apoptosis) with cell death itself, as several recent studies have indicated.[9,10]
4. Better attention to the effects of chemoradiation on normal tissues, including the acute toxicities that can limit dose intensity and the late toxicities that can impair the quality of life of the cured patient. Are the effects on normal tissue greater than that seen with either modality alone or with both modalities administered in sequence?

While treatment with both local and systemic modalities for a number of adult solid tumors will improve survival or local control and organ preservation, the need for concurrent therapy has not been shown in these sites. Although there are appealing rationales for concurrent as opposed to sequential therapy from both the laboratory and the clinic, these have been imperfect guides and require careful scrutiny in the form of prospective trials before such strategies can become accepted as standard therapy.

The integration of the three major modalities of cancer treatment in the 1990s - surgery, radiation therapy, and chemotherapy - produces effects that have clinical, economic, and psychosocial dimensions. While we might agree in principle that a patient treated with concurrent chemoradiotherapy should be followed by both radiation and medical oncologists, in practice this rarely happens. With the emerging dominance of primary care gatekeepers in medical care, it is unclear than oncologic specialists will be allowed (ie, reimbursed for) assessment of the consequences of their treatments, for good or ill. Yet, complex multimodality regimens require careful management of dosage, scheduling, and toxicities, which typically requires the coordination of several different specialists. Without this coordination, we risk either expensive duplication of care and tests or undermanagement of the patient when each physician assumes that the other is managing the patient's care (eg, monitoring blood counts and checking the results of the brain magnetic resonance images). In these situations, oncologic nurse specialists who work with both the radiation oncologists and the medical oncologists can provide a valuable linkage. It also is an area in which the development of clinical guidelines for staging, treatment, and outcome assessment is a mandate.

None of these caveats detracts from the clinical observation that combinations of radiation and chemotherapy often produce dramatic clinical responses and provide effective therapy for malignancies that are refractory to either modality. Prospective, randomized trials in several disease sites, including the rectum, esophagus, lung, and breast, show that the use of both modalities is superior to either used separately. It is less clear that the concurrent rather than the sequential use of both modalities is needed. We must be cautious in our presumption that the clinical gain is derived from some specific interactions rather than simply from the use of both modalities of treatment in a short overall time.

Rather than a marriage, a time-sharing arrangement may be a better metaphor for the combined use of radiation and chemotherapy, with interactions minimized rather than maximized.

Henry Wagner, Jr, MD

Program Leader
Thoracic Oncology Program
H. Lee Moffitt Cancer Center & Research Institute
Associate Professor of Radiology
University of South Florida
Tampa, Florida

References

1. Steel G. Terminology in the description of drug-radiation interactions. Int J Radiat Oncol Biol Phys. 1979;5:1145-1150.
2. Nachman J, Sather H, Lukens S, et al. Cranial radiation (CRT) improves event free survival (EFS) for high risk patients with acute lymphoblastic leukemia (ALL) showing a rapid response (RR) to BFM induction chemotherapy. Proc Annu Meet Am Soc Clin Oncol. 1994;13:A1042.
3. Minarik L, Hall EJ. Taxol in combination with acute and low dose rate irradiation. Radiother Oncol. 1994;32:124-128.
4. Steel GG. Cell synchronization unfortunately may not benefit cancer therapy. Radiother Oncol. 1994;32:95-97.
5. Choy H, Browne MJ. Paclitaxel as a radiation sensitizer in non-small lung cancer. Semin Oncol. 1995;22:70-74.
6. Garcia de Palazzo I, Adams GP, Sundareshan P, et al. Expression of mutated epidermal growth factor receptor by non-small cell lung carcinomas. Cancer Res. 1993;53:3217-3220.
7. Abrams SI, Hand PH, Tsang KY, et al. Mutant ras epitopes as targets for cancer vaccines. Semin Oncol. 1996;23:118-134.
8. Bernhard EJ, Kao G, Cox AD, et al. The farnesyltransferase inhibitor 227 is a radiosensitizer of cells expressing activated H-ras. Proc Am Assoc Cancer Res. 1996;37:604. Abstract 4141.
9. Rudoltz MR, Bernhard EJ, Kao GD, et al. Apoptosis, cell cycle kinetics, and clonogenic survival in a solid tumor model system expressing bcl-2. Proc Am Assoc Cancer Res. 1996;37:602. Abstract 4126.
10. Zhen W, Loviscek K, Walter S, et al. Altered access to the radiation-induced apoptotic pathway in TK6 cells does not affect clonogenic survival after irradiation. Proc Am Assoc Cancer Res. 1996;37:602. Abstact 4127.

Back to Cancer Control Journal Volume 3 Number 4