4 Principles of Combining Radiation
Therapy and Chemotherapy
LUKA MILAS AND JAMES D. COX
THERAPEUTIC RATIO SEQUENCING OF CHEMOTHERAPY
AND RADIATION THERAPY
IMPROVING THE THERAPEUTIC RATIO
Induction Chemotherapy
Spatial Cooperation
Concurrent Chemotherapy
Independence of Toxicity
Adjuvant Chemotherapy
Enhancement of Tumor Response
CLINICAL APPLICATIONS OF
Protection of Normal Tissues
COMBINED RADIATION THERAPY
ADDITIVITY OF DRUG-RADIATION
AND CHEMOTHERAPY
INTERACTIONS
SEQUENTIAL CHEMOTHERAPY
METHODS FOR ASSESSING
AND RADIATION THERAPY
DRUG-RADIATION EFFECTS
Induction Chemotherapy
MECHANISMS OF DRUG-RADIATION
Adjuvant Chemotherapy
INTERACTIONS
Concurrent Chemotherapy and Radiation
Initial Radiation Damage
Therapy
Inhibition of Cellular Repair
Chemoradiation plus Surgery
Cell Cycle Effects
Organ Conservation
Hypoxia
Toxicity Issues
Cell Regeneration
Inhibition of Tumor Angiogenesis
In perhaps no other aspect of oncologic practice has a which is defined as the ratio of the maximally tolerated
direction developed that is as clear and compelling as drug (in this case, with regard to normal tissue damage,
the use of chemotherapy in conjunction with radiation curve C) to the minimally curative or effective dose (in
therapy or radiation therapy and surgery. The question of this case, the antitumor effect, curves A or B). Any level of
whether to use combined-modality treatment has changed probability can be used to calculate the therapeutic ratio;
to which sequence and which drugs are the best. Given the however, in experimental preclinical studies, 50% is the
increasing availability of preclinical investigations that can most commonly used.
be used to chart a course for combining radiation therapy The objective of any cancer treatment is to achieve a
with drugs, it is important to explore in some depth the positive therapeutic ratio (i.e., one in which the antitumor
preclinical findings that have pointed the way to the use efficacy curve lies to the left of the curve representing nor-
of combined-modality therapy. It is equally important to mal tissue injury [see Fig. 4-1, curve A]). Treatments for
clarify the essential steps by which clinical investigations which the therapeutic ratio is negative (i.e., the antitumor
establish the toxicity, efficacy, and proper place for response curve lies to the right of that for normal tissue
chemotherapy in combined-modality therapies in standard damage [see Fig. 4-1, curve B], and normal tissue damage
practice. In this chapter, we define some basic concepts, is more likely than tumor damage) are contraindicated,
outline mechanisms of interaction between irradiation and except perhaps for palliation in some circumstances.
chemotherapy, present theoretical principles underlying In defining therapeutic ratio, the end points of tumor re-
sequencing strategies, and review clinical applications of sponse and normal tissue injury must be appropriately cat-
the combined-treatment strategies. egorized and quantified. The end points depend on factors
such as whether the treatment is intended to be curative
or palliative and which tissues limit the doses that can be
THERAPEUTIC RATIO
given. Damage to normal tissue must be within acceptable
Neither radiation nor chemotherapeutic drugs are parti- limits in terms of the tissue type and the severity of the
cularly selective against tumors; both also damage normal damage. The balance between complications arising from
tissues. This damage limits the dose of radiation or drugs, normal tissue damage and the probability of tumor control
or both, that can be given to patients. To be therapeutically form the basis of the therapeutic ratio of a treatment.
beneficial, individual agents or their combination must
be more effective against tumors than against normal
IMPROVING THE THERAPEUTIC RATIO
tissues. Both antitumor efficacy and normal tissue injury
are positively related to the dose of radiation or cytotoxic The ultimate goal in combining chemotherapeutic drugs
drugs, and this relationship is commonly described by with radiation therapy is to improve the therapeutic
a sigmoid curve (Fig. 4-1). The plots shown in Figure ratio to enhance the antitumor effect while minimizing
4-1 provide information on the therapeutic ratio or index, the toxicity of the treatment to critical normal tissues. Steel
102
4 Principles of Combining Radiation Therapy and Chemotherapy 103
A C B detail later in this chapter. Virtually all chemotherapeutic
100
agents enhance radiation damage to normal tissues as well
as malignancies; consequently, therapeutic benefit can be
achieved if the susceptibility of tumors rather than that of
75
critical normal tissues can be enhanced.
Protection of Normal Tissues
50
Some compounds are thought to increase the tolerance of
normal tissues to radiation so that higher doses of radiation
can be safely delivered to the tumor. Preclinical in vivo
25
testing has shown that some chemicals can selectively or
preferentially protect normal tissues.2-4 Among the commonly
tested protectors are thiol compounds such as amitostine
0
(WR-272), which may be useful in clinical settings, such as
Dose
protecting salivary glands during radiation therapy delivered
to the head and neck.5 A major problem with concurrent
FIGURE 4-1. The therapeutic ratio is the ratio of the doses of an
agent or agents that produces the same probability of normal chemoradiation therapy is the increase in the toxic
tissue damage (curve C) and antitumor effect (curves A and B).
effects on tissues that is conferred by radiation. However,
Curve A illustrates an agent with a positive (beneficial) thera-
radioprotectants and technical improvements in radiation
peutic ratio, one that produces more damage to tumor than to
therapy, such as three-dimensional treatment planning,
normal tissue; curve B illustrates a negative (undesirable) thera-
conformational radiation therapy, or the use of protons, are
peutic ratio in which the damage to normal tissue exceeds the
likely to minimize tissue toxicity and consequently enhance
damage to the tumor.
the effectiveness of chemoradiation.
ADDITIVITY OF DRUG-RADIATION
and Peckham1 defined four principles by which therapeutic
INTERACTIONS
ratios of combined treatments can be improved: spatial
cooperation, independence of toxicity, enhancement of Interactions between chemotherapeutic agents and radia-
tumor response, and protection of normal tissues. tion may result in additive, supra-additive, or subadditive
effects, depending on whether the combination produces
Spatial Cooperation
effects that are equal to, greater than, or less than,
Treatments that act independently of each other at different respectively, the sum of the activity produced by the
anatomic sites are said to exhibit spatial coordination. For individual agents. Evaluations of the additivity of a given
example, chemotherapy is commonly used to deal with combination must take into account the dose-response
tumor growth outside the radiation field, usually to eradicate relationship for each individual agent. The calculations
metastases. Conversely, radiation can be used to treat are simple when both agents have effects that are
neoplastic lesions in secluded sites that are poorly accessed exponentially related to the dose. However, the dose-effect
by chemotherapeutic agents; an example of this strategy relationship is rarely exponential. As described in Chapter 1,
is the use of irradiation to supplement chemotherapeutic many radiation dose-cell survival curves have a shoulder
treatment of leukemia that has spread to the brain. of varying widths at lower doses of irradiation and an
exponential relationship at higher doses. The presence of
Independence of Toxicity
such a shoulder indicates that the cells can repair radiation
Because toxicity to normal tissues is the principal dose- damage. The dose-survival curves after chemotherapy tend
limiting factor for chemotherapy and for radiation therapy, to be more diverse; some have shoulders and some do not,
the antitumor efficacy of a combined treatment theoretically and some show tails at higher drug doses. The presence
would be improved more readily and the treatment tolerated of such tails implies the existence of cell subpopulations
more easily through the use of drugs with toxic effects that are resistant to the chemotherapeutic agents. These
that do not overlap or enhance those induced by radiation nonlinear, dose-related characteristics in cell killing imply
therapy. Use of this principle requires thorough knowledge that when two agents are combined, additive effects can
of the toxicity, pharmacology, and optimal administration of be produced over a range of doses.
individual chemotherapeutic drugs to ensure that combining Determinations of additivity on the basis of a combi-
such drugs with radiation minimizes normal tissue damage nation of only one dose of each agent are inappropriate
while still retaining antitumor efficacy. and can be misleading. To overcome this problem, Steel
and Peckham1 introduced the concept of isobologram
Enhancement of Tumor Response
analysis, in which an isoeffect plot is generated for the
Some drugs can enhance radiation-induced tumor damage, dose-response pattern for the combination of two agents
or radiation can make some cells more sensitive to (Fig. 4-2). Dose-response curves are generated for each
chemotherapy, presumably through interactions between agent, and the curves are used to generate isobolograms,
the agents at the molecular, cellular, or pathophysiologic which illustrate the expected additive response. In the sim-
(microenvironmental or metabolic) levels. Some of the plest situation, that in which each agent produces a linear
more important mechanisms of drug-radiation interactions dose response, the isobologram is also linear (see Fig. 4-2A).
that increase tumor susceptibility are considered in more If the combination of the two agents has a supra-additive
Probability of tumor control
or normal tissue damage (%)
104 PART 1 Principles
A B C
Subadditive
or protection
Supra-additive
Radiation dose (Gy)
or synergy
FIGURE 4-3. The effects of chemotherapeutic drugs on cell sur-
vival after irradiation are shown by blue lines indicating radiation
only and by red indicating drug plus radiation. A, Simple
displacement of radiation survival curve indicates additivity.
A
B, Elimination of the shoulder indicates inhibition of sublethal
damage repair. C, Change in the slope of the radiation curve
indicates sensitization (displacement to the left) or protection
(displacement to the right; green line).
Subadditive
or protection
survival or clonogenic assays, which involve measuring the
ability of cells to survive and produce colonies of progeny of
a defined minimum size (i.e., ability to sustain reproductive
integrity). Survival curves are usually designed by plotting
the surviving fraction of cells on a logarithmic scale against
the dose of radiation or drugs, which is plotted on a linear
scale (Fig. 4-3). Established cell lines typically are used to
Supra-additive
assess clonogenic survival after treatment with drugs or
or synergy
radiation in vitro, although a variation of the clonogenic
B Dose of agent B
cell survival assay called the excision assay can also be used
to study antitumor treatments given in vivo. In an excision
FIGURE 4-2. Isobolograms are constructed for pairs of agents
that have linear dose-response curves (A) or nonlinear dose- assay, the tumor is excised at some specified time after
response curves (B). The envelope of additivity and regions of treatment, and a single-cell suspension of tumor cells is
supra-additivity and subadditivity also are shown. (Modified
prepared and plated for subsequent determination of cell
from Steel GG, Peckham MJ. Exploitable mechanisms in com-
survival.
bined radiotherapy-chemotherapy: the concept of additivity. Int
The addition of drugs to radiation can influence sur-
J Radiat Oncol Biol Phys 1979;5:85-91.)
vival curves in several ways (see Fig. 4-3). First, the drugs
may shift the curve downward without changing its shape.
The magnitude of this displacement corresponds to the
effect, the response appears to the left of the curve; if the amount of cells killed by the drugs (see Fig. 4-3A). Sec-
combination has a subadditive effect, the response appears ond, the drugs may eliminate the shoulder of a radiation
to the right of the curve. When the dose response is non- survival curve, which implies that the drugs are inhibiting
linear, an envelope of additivity appears in the isobologram the cellular radiation-repair process (see Fig. 4-3B). Third,
(see Fig. 4-2B), indicating a range of isoeffect curves that drugs may change the slope of the exponential portion of
represent additive responses. In this situation, to obtain an the survival curve (see Fig. 4-3C); in this case, a steeper
additive effect, the doses of individual agents that produce slope indicates sensitization and a shallower slope indicates
the same biologic effect lie within the envelope. If the dos- protection.
es lie to the left of the isobologram envelope, the interac- Many methods are available for assessing the effects
tion is supra-additive or synergistic the effect is caused of chemotherapeutic drugs on tumors and normal tissues
by doses of the two agents that are lower than would be in vivo.6,7 Two of the most common assays used to assess
predicted from the envelope of additivity. In contrast, if tumor response are the tumor growth delay assay8 and the
the doses lie to the right side of the envelope, the inter- tumor-control dose (TCD50) assay.9 When the treatment
action is subadditive or antagonistic, meaning the effect end point is a delay in tumor growth, the size of untreat-
requires higher doses of the two agents than predicted. The ed and treated tumors is measured as a function of time
envelope of additivity becomes wider as the gap between and plotted in growth curves. The absolute growth delay
the nonlinear dose responses to individual agents becomes represents the difference in time (usually in days) needed
greater. for untreated and treated tumors to reach a defined size
(Fig. 4-4A). When a drug is given with radiation, the growth
delay achieved by both modalities minus the delay caused
by the drug alone is called the normalized growth delay.
METHODS FOR ASSESSING
To determine whether the addition of drugs enhances or
DRUG-RADIATION EFFECTS
reduces the antitumor efficacy of radiation requires deter-
Many in vitro and in vivo assays are used to assess the mining normalized growth delay after at least three differ-
biologic effects of drugs, radiation, and the combination of ent doses of radiation. The normalized growth delays after
both.6,7 In vitro assessments most commonly involve cell the combined treatment and the absolute growth delays
Cell survival
Dose of agent A
En
v
elope of additivity
4 Principles of Combining Radiation Therapy and Chemotherapy 105
Control Drug Xrt Drug+xrt enhancement of the radiation response of a mouse mam-
mary carcinoma by the administration of docetaxel before
irradiation.10
In the TCD50 assay, the end point is tumor control or
cure in 50% of the treated subjects. In this assay, the pro-
portion of tumors cured by radiation only or by a drug-plus-
radiation combination is plotted as a function of radiation
dose. If the combination displaces the radiation-only curve
to the left, addition of the drug improves tumor curability;
however, the displacement does not discriminate between
improvement caused by independent cell kill by the drug
and that caused by a supra-additive interaction between
the drug and radiation.
Many assays can be used to quantify the radiation dose-
response relationship for normal tissues and the influ-
0 A B C D
ence of chemotherapeutic drugs on the radiation response
Time after treatment
A (Table 4-1). Some of these assays are clonogenic; the end
point depends directly on the reproductive integrity of
individual cells. More frequently, however, dose-response
20
relationships for normal tissues are based on functional
18
end points. These end points tend to reflect the minimum
number of functional cells remaining in tissues or organs
16
rather than the proportion of cells that retain reproductive
integrity.
14
12
MECHANISMS OF DRUG-RADIATION
10
INTERACTIONS
8
Clinical experience has shown that chemotherapy is
most effective in improving local tumor control when it
6
is given during the course of radiation therapy. This obser-
vation implies that chemotherapeutic drugs and radiation
4
actively interact. The multiple mechanisms underlying
that interaction, which occur at the cellular and tumor-
2
microenvironment levels, are complex; six major mech-
8 10 12 14 16 18 20 22
anisms are considered in the following paragraphs.
B Radiation dose (Gy)
FIGURE 4-4. The graph shows the delay in tumor growth af- Initial Radiation Damage
ter treatment with a drug, radiation, or a combination of both.
The critical target for radiation injury is the DNA molecule
A, The points at B, C, and D are absolute growth delays after
(see Chapter 1). Radiation induces several types of lesions,
treatment with a drug (red line), radiation (Xrt; green line), or
including single-strand breaks, double-strand breaks, base
drug plus radiation (Drug + xrt; yellow line). Normalized growth
damage, and DNA-DNA or DNA-protein cross-links.
delay is defined as D minus B. B, The interactions between the
The most common type of lesion is the single-strand
chemotherapeutic drug docetaxel and radiation in a murine
break, which is readily repaired. Double-strand breaks left
mammary carcinoma are plotted. Addition of the drug displaced
the radiation response curve to the left, indicating enhancement unrepaired or misrepaired and associated chromosomal
of the tumor radioresponse. Green circles represent the tumor
aberrations usually lead to cell death.11 Certain drugs, such
response after radiation only; red circles and blue triangles
as halogenated pyrimidines, that become incorporated
represent the tumor radioresponse when docetaxel was given
into DNA make the DNA more susceptible to radiation
9 or 48 hours before irradiation, respectively. (From Mason KA,
damage.12 In addition to increasing the initial injury
Hunter NR, Milas M, et al. Docetaxel enhances tumor radiore-
caused by radiation, halogenated pyrimidines can enhance
sponse in vivo. Clin Cancer Res 1997;3:2431-2438.)
cell radiosensitivity by interfering with cellular repair
mechanisms.12,13
Inhibition of Cellular Repair
after irradiation only are plotted as a function of radiation
dose (see Fig. 4-4B). The displacement of the normalized Cells have the ability to repair sublethal14 and potentially
growth delay curve in relation to the absolute growth delay lethal15 radiation damage. Both types of repair are defined
curve indicates the effect of drugs on tumor radiorespon- in operational terms. Sublethal damage repair is defined as
siveness. No displacement indicates a simple additive effect; the increase in survival observed when a dose of radiation
a shift to the left (to lower radiation doses) indicates en- is split into two (or more) fractions with some time interval
hancement of tumor radioresponse; and a shift to the right between. Sublethal damage repair manifests on cell
(to higher radiation doses) indicates a reduction in tumor survival curves as restoration of a shoulder on the curve
radioresponse. The results shown in Figure 4-4B illustrate after the second dose. Repair takes place quickly, typically
Tumor size
Normalized growth delay (days)
106 PART 1 Principles
TABLE 4-1. In Vivo Assays for Normal Tissue Damage by Drug-Radiation Combinations
Tissue Clonogenic End Points Functional End Points
Bone marrow Exogenous and endogenous spleen LD50, depletion of different cell types
colony assay
Skin In situ skin colonies Skin erythema and desquamation, hair loss,
telangiectasia, skin contraction
Gastrointestinal tract Jejunum and colon microcolonies LD50, weight loss, protein or electrolyte leakage
Lung LD50, breathing rate, CO uptake
Kidney Kidney tubules Urine output
Bladder Urination frequency
Testis Spermatogonial stem cell assay Testis weight, sperm count, fertility
Central nervous system LD50 (brain), paralysis (spinal cord)
CO, carbon monoxide; LD50, lethal dose in 50%.
with a half-time of about 1 hour and a completion time of
TABLE 4-2. Radiation-Induced DNA Damage and
4 to 6 hours after irradiation, and it takes place in tumors,
its Repair
normal tissues, and cultured cells. Potentially lethal damage
Induced Lesions Types of Repair
repair designates an increase in the proportion of cells
that survived irradiation because of some modification of
Single-strand breaks Direct reversal
postirradiation conditions. Potentially lethal damage repair
Double-strand breaks Homologous and nonhomologous
typically takes place under conditions that prevent cellular
recombination
division for several hours, such as maintaining cultured
Base alterations Base excision repair
cells in a plateau phase after irradiation or delaying the
removal of cells from tumors for clonogenic survival assay.
DNA-protein cross-links Nucleotide excision repair
Conceptually, repair of potentially lethal damage can be
DNA, deoxyribonucleic acid.
explained in terms of successful repair of DNA lesions
that would have been lethal had the DNA undergone
replication within several hours of irradiation. Repair
can be achieved through several mechanisms, such as phosphorylation by deoxycytidine kinase to produce the
restoration of damaged molecules by reducing species that active metabolites gemcitabine diphosphate (dFdCDP)
donate electrons to oxidized substrates and through the and gemcitabine triphosphate (dFdCTP).17,18,20 Compared
involvement of a variety of repair enzymes. Possible repair with fludarabine, gemcitabine has greater membrane per-
mechanisms for the various DNA lesions are summarized meability and enzyme affinity and has longer intracellular
in Table 4-2. retention.17,20 Gemcitabine diphosphate interferes with
Many chemotherapeutic drugs can interact with cellu- DNA synthesis by inhibiting ribonucleotide reductase and
lar repair mechanisms, and in inhibiting those mechanisms, reducing the deoxynucleotide pools. Gemcitabine triphos-
the drugs may enhance the response of cells or tissues phate competes with deoxycytidine triphosphate for incor-
to radiation. Halogenated pyrimidines can enhance cell poration into elongated DNA strands; only one additional
radiosensitivity by increasing initial radiation damage and deoxynucleotide can be incorporated after the insertion of
by inhibiting cellular repair. Significant effort has focused dFdCTP, halting DNA polymerization.17,20 In contrast to
on exploring the radioenhancing effects of nucleoside ana- fludarabine, gemcitabine is active against many common
logues such as fludarabine16,17 and gemcitabine16-18 that solid tumors in humans, including head and neck, lung, and
can inhibit the repair of radiation-induced DNA or chro- pancreatic carcinoma.16,17
mosome damage. Fludarabine is a potent inhibitor of DNA Nucleoside analogues such as fludarabine and gemcitabine
primase, DNA polymerase-Ä…, µ-DNA ligase, and ribonucle- typically are potent radiosensitizers of mammalian cells
otide reductase, and it can cause DNA chain termination.19 in vitro16-18,21 and enhancers of tumor radioresponse in
Fludarabine phosphate is rapidly dephosphorylated in vivo vivo.16,17,22-24 Investigations from our laboratory have dem-
to form fludarabine, an adenine nucleoside analogue that onstrated that fludarabine and gemcitabine can significantly
is transported by a carrier into cells, where it is converted enhance the radioresponse of several mouse solid tumors,
to the active metabolite by the addition of adenosine tri- as assessed by tumor growth delay or local tumor cure,
phosphate. Transport into cells and metabolic activation of after single-dose or fractionated-dose irradiation.16,17,22-24
fludarabine seem to vary among cell types, with malignant Fludarabine in particular produced radioenhancement
cells concentrating and metabolizing the drug more effi- ranging from a factor of 1.2 to more than 2.0, depend-
ciently than normal cells do. Fludarabine is clinically active ing on the timing of fludarabine administration relative to
against leukemias and lymphomas, but it is not particularly irradiation, the dose of fludarabine, the type of tumor, and
effective against solid tumors. schedule of irradiation. Although several mechanisms seem
Gemcitabine (22 -deoxy-22 ,22 -difluorocytidine) is a to have been responsible for the enhancement, inhibition
fluorine-substituted cytarabine that requires intracellular of radiation damage repair seems to have been prominent.
4 Principles of Combining Radiation Therapy and Chemotherapy 107
100
100
Single dose
80
Control
10-1
60
40
1 nM
10 nM paclitaxel
paclitaxel
10-2
20
0
10-3
0 2 4 6 8
Absorbed dose (Gy)
100
Fractionated
FIGURE 4-6. Survival curves are plotted for astrocytoma cells
treated with radiation and paclitaxel. Survival is presented at a
particular radiation dose and expressed relative to that of nonir-
80
radiated controls given the same dose of paclitaxel. Points are
means Ä… standard error. Symbols represent untreated control
cells (blue circles), cells treated with dimethylsulfoxide (red cir-
60
cles), 1 nM paclitaxel (open green triangles), or 10 nM paclitaxel
(closed green triangles). (From Tishler RB, Geard CR, Hall EJ, et
al. Taxol sensitizes human astrocytoma cells to radiation. Cancer
40
Res 1992;52:3495-3497.)
20
was first reported in 1963 by Terasima and Tolmach.25
In general, cells in the G2 and M phases are the most
sensitive, and cells in the S phase are the most resistant.
0
This variation in radiosensitivity over the cell cycle can be
20 40 60 80 100 120 exploited for designing effective chemoradiation therapy
strategies such as the use of drugs that accumulate cells in
Total radiation dose (Gy)
a radiosensitive phase or eliminate radioresistant S-phase
FIGURE 4-5. Fludarabine-induced enhancement of the
cells. Two examples can illustrate these concepts.
response of the FSA mouse fibrosarcoma line to single-dose or
Taxanes, of which paclitaxel and docetaxel are proto-
fractionated-dose (16 fractions with a 6-hour interval between
types, are potent mitotic spindle poisons. Taxanes inhibit
fractions) irradiation. Fludarabine (400 mg/kg) was given
tubulin depolymerization by binding to ²-tubulin, increas-
3 hours before single-dose irradiation (A) or daily during the
4-day fractionated treatment (B) starting 3 hours before the first ing tubulin polymerization and promoting microtubule
fractional dose of radiation. Blue lines indicate radiation only;
assembly and stability, which collectively leads to the cells
red lines indicate radiation plus fludarabine. (From Gregoire V,
becoming arrested in the radiosensitive G2 and M phases of
Hunter NR, Brock WA, et al. Improvement in the therapeutic ra-
the cell cycle.26 The ability of taxanes to block cells in G2/M
tio of radiotherapy for a murine sarcoma by indomethacin plus
was the biologic rationale that led to extensive testing of
fludarabine. Radiat Res 1996;146:548-553.)
their radioenhancing properties. In the first report on the
radioenhancing effect of taxanes,27 the radiosensitivity of a
The effect of fludarabine on the radiocurability of a murine human astrocytoma cell line was reportedly increased by a
sarcoma after single-dose or fractionated irradiation is illus- factor of 1.8 (Fig. 4-6). Radiosensitization was achieved only
trated in Figure 4-5. The enhancement was much greater when the cells were in phase G2 or M at the time of irradia-
when fludarabine was combined with fractionated irradia- tion, a result subsequently confirmed in many in vitro and
tion, suggesting that the inhibition of sublethal (or poten- in vivo studies.28 Cells arrested in mitosis by taxanes often
tially lethal) damage repair was a significant mechanism of die by apoptosis (the underlying basis for the antitumor
enhancement of tumor radioresponse. efficacy of these drugs), but they may overcome the mi-
totic block and continue with division. In that case, tumors
Cell Cycle Effects
are resistant to treatment with taxanes alone, but they can
The cytotoxicity of most chemotherapeutic agents and show an enhanced radioresponse if radiation is delivered at
ionizing radiation depends greatly on the proliferative state the time of mitotic arrest, usually between 6 and 12 hours
of the cells, with most agents being more effective against after administration of the drug. (Tumor reoxygenation,
proliferating cells than nonproliferating cells. Sensitivity to another major mechanism of radioenhancement in tumors
such agents also varies according to the phase of the cell that respond to taxanes by mitotic arrest and apoptosis, is
cycle. The influence of cell cycle in response to radiation discussed in more detail under  Hypoxia. ) The histologic
Relative survival
Tumor cure (%)
108 PART 1 Principles
Chemotherapeutic agents can improve the antitumor
efficacy of radiation therapy by reducing or eliminating the
negative (tumor-protective) effects of hypoxic cells. This
improvement can be achieved by (1) eliminating tumor
cells in oxygenated regions, (2) selectively eliminating
hypoxic cells, or (3) sensitizing hypoxic cells to radiation.
Elimination of cells in oxygenated regions leads to tumor
reoxygenation (see Chapter 1), and most chemotherapeu-
tic drugs can achieve this effect because they preferentially
kill well-oxygenated, proliferating cells. Eliminating oxy-
genated cells makes more oxygen available to the surviving
cells by lowering the interstitial pressure on microvessels
within the tumor, thereby reopening closed capillaries and
increasing the intratumoral blood supply. Tumor shrinkage
from cell loss and active migration of tumor cells also brings
FIGURE 4-7. The histologic appearance of mouse MCa-4 previously hypoxic microregions closer to blood vessels.
mammary tumors treated with 33 mg/kg of docetaxel features
Tumor reoxygenation has been shown to be a major
mitotically arrested cells (short arrows) and apoptotic cells
mechanism by which taxanes enhance tumor radiorespon-
(long arrows) 12 hours after treatment with docetaxel. Mitoti-
siveness (Fig. 4-8).30 In that study, giving paclitaxel to mice
cally arrested cells show the characteristic coronal appearance
bearing an adenocarcinoma within 3 days before local irra-
(×1000).
diation enhanced the radiosensitivity of the tumors. How-
ever, this enhancing effect was reduced when the tumors
appearance of mitotically arrested and apoptotic cells in were rendered hypoxic by clamping the tumor-bearing legs
murine MCa-4 tumors treated with docetaxel is shown in at the time of irradiation. Paclitaxel s ability to increase
Figure 4-7. tumor oxygenation was confirmed directly by measuring
Another approach for exploiting the cell cycle effect in the intratumoral partial pressure of oxygen (PO2).30
chemoradiation therapy is to use drugs that eliminate ra- Another chemotherapeutic approach that increases
dioresistant S-phase cells. In addition to inhibiting radiation tumor radiosensitivity is to use drugs that selectively kill
damage repair, nucleoside analogues such as fludarabine or hypoxic cells. Several drugs with this property are available,
gemcitabine become incorporated into S-phase cells, many including tirapazamine and mitomycin31; their cytotoxicity
of which then die by apoptosis.5,17,24,29 This preferential results from their reductive activation under hypoxic con-
removal of the S-phase component from tumors leads to ditions. Some active metabolites can diffuse into surround-
a parasynchronous movement of surviving cells through ing oxygenated regions and kill oxygenated cells. Hypoxia
S phase and mitosis, resulting in accumulation of cells in is associated with other metabolic alterations, including the
the G2 and M phases of the cell cycle between 24 and development of low pH from the production of lactic acid
36 hours after administration of the drugs, when the high- through hypoxia-driven anaerobic metabolism.32 The acid-
est radioenhancement values are observed. Elimination of ity is extracellular and does not seem to affect intracellular
radioresistant S-phase cells from the tumor cell population pH.32 The acidic state of tumors can be therapeutically ex-
and redistribution of surviving cells into more radiosensitive ploited by using drugs that selectively accumulate in acidic
compartments of the cell cycle seem to be involved in the compartments or become activated at low pH values.33
radioenhancement. Another approach to reducing the negative influence of
hypoxic cells is to use drugs that selectively radiosensitize
Hypoxia
the cells.34 These drugs mimic the effect of oxygen in in-
The presence of oxygen has long been known to influence cell creasing radiation damage. Misonidazole and other hypoxic
radiosensitivity; hypoxic cells are 2.5 to 3 times less sensitive cell radiosensitizers are highly effective in enhancing the
to radiation than are well-oxygenated cells. Because tumors radioresponsiveness of tumors in rodents. Among the many
often include hypoxic areas but normal tissues usually clinical trials conducted, only a few have shown improved
do not, the radioresistant effect of hypoxia typically is outcome from the combination of hypoxic-cell radiosensi-
restricted to tumors. Blood circulation within solid tumors tizers with radiation therapy. The major limitation of these
tends to be poor because of abnormalities in the number, combinations in humans has been neurotoxicity, which
structure, and function of blood vessels within the tumor. precludes the use of effective doses of the hypoxic-cell
Tumor blood vessels are commonly irregular and tortuous, radiosensitizers.
and they have blind ends, arteriovenous shunts, and
Cell Regeneration
incomplete endothelial linings and basement membranes.
All of these factors can limit intratumoral blood flow In normal adult tissues, a balance is maintained between
and the delivery of oxygen and nutrients, resulting in cell production and cell loss such that no net growth
many tumor microregions becoming hypoxic, acidic, and occurs. Radiation or chemotherapeutic drugs alter this
eventually necrotic. Hypoxia develops in tissues when the balance by increasing cell loss; to restore the balance, the
distance from blood vessels exceeds about 100 to 150 µm; rate of cell production among the surviving cells increases.
the proportion of hypoxic cells in tumors can often exceed In clinical radiation therapy, which is typically given over
50%. Cells in hypoxic regions stop proliferating. In contrast, several weeks, the cell loss that occurs after each fractional
cells that are nearer to blood vessels are well oxygenated dose is counteracted by the compensatory cell regeneration
and actively proliferate. (i.e., repopulation) that takes place between fractions. The
4 Principles of Combining Radiation Therapy and Chemotherapy 109
80
100
Air Control
60 Gy-primed
70
80
60
60
50
40
40
20
30
20
0
0 7 14 21
A
Days
100
Hypoxia FIGURE 4-9. Values for the radiation dose yielding local
tumor control in 50% of the animals tested (TCD50) for mouse
MCa-4 tumors are plotted with 95% confidence intervals as a
function of time after tumor cell inoculation (blue squares) or
80
after 8-mm established tumors had been exposed to 60 Gy
(red squares). The linear increase in TCD50 is consistent with
exponential growth, which was significantly faster in the pre-
60
irradiated tumors. (From Milas L, Yamada S, Hunter N, et al.
Changes in TCD50 as a measure of clonogen doubling time in
irradiated and unirradiated tumors. Int J Radiat Oncol Biol Phys
1991;21:1195-1202.)
40
measured by changes in TCD50 values. In this study, estab-
lished 8-mm murine mammary carcinomas were irradiated
20
with a single dose of 60 Gy, which allowed only a small
number of tumor clonogens to survive. At different days
thereafter, while the tumors regressed, transiently disap-
0
peared, or regrew, the tumors were irradiated again with
graded doses to establish TCD50 values. The later doses were
30 40 50 60 70 80 90
given under hypoxic conditions to exclude the influence of
oxygen on tumor radioresponse. The TCD50 values of tu-
Radiation dose (Gy)
B
mors, regardless of size, that received no priming radiation
FIGURE 4-8. The effect of paclitaxel on radiation dose-response
dose served as controls. As shown in Figure 4-9, the slope of
curves is plotted for local control of murine MCa-4 tumors. Mice
the TCD50 curve for the 60-Gy primed tumors was signifi-
were irradiated under oxygenated (A) or hypoxic (B) conditions.
cantly steeper than that for the control tumors, implying
Horizontal lines represent 95% confidence limits at the radiation
an increased rate of clonogen proliferation in the irradiated
dose yielding local tumor control in 50% of the animals tested
(TCD50). (From Milas L, Hunter NR, Mason KA, et al. Role of reox- tumors. A similar accelerated repopulation was observed in
ygenation in induction of enhancement of tumor radioresponse murine tumors treated with cyclophosphamide.39
by paclitaxel. Cancer Res 1995;55:3564-3568.)
Accelerated proliferation of tumor clonogens seems to
occur during clinical radiation therapy. Withers and associ-
extent of cell repopulation determines the tolerance of ates40 reviewed the relationships among the probability of
acutely responding tissues to radiation. achieving local control of head and neck carcinoma, total
In malignant tumors, the balance between cell pro- radiation dose, and treatment duration. In that study, the
duction and cell loss is shifted in favor of cell production, TCD50 increased progressively over time if the treatment
resulting in uncontrolled growth. In tumors and in normal period was prolonged beyond 30 days (Fig. 4-10). The in-
tissues, radiation and other cytotoxic treatments produce crease in curative dose was greater than anticipated from
massive cell loss, which stimulates a compensatory regen- estimates of pretreatment tumor volume and clonogen dou-
erative response. Studies of cell regeneration in experimen- bling rates. In the study, the dose increase needed to offset
tal tumors after single and fractionated radiation doses have accelerated clonogen repopulation was about 0.6 Gy/day;
shown that the rate of cell proliferation usually is higher however, that increase can exceed 1 Gy/day.41
in treated tumors than in untreated tumors, a phenome- Acutely responding normal tissues and tumors respond to
non called accelerated cell proliferation.35-38 An example of radiation- or chemotherapy-induced injury with increased
accelerated proliferation after tumor irradiation is shown clonogen production. Although accelerated cell prolifera-
in Figure 4-9, for which the rate of cell proliferation was tion can be beneficial in terms of sparing normal tissues,
50
TCD
(Gy)
Tumor cure (%)
Tumor cure (%)
110 PART 1 Principles
for the failure of induction chemotherapy to improve radia-
tion therapy. One possibility is that response to chemother-
apy predicts the response to radiation therapy. According
70
to this reasoning, tumor cells that are sensitive or resistant
to chemotherapeutic drugs are also sensitive or resistant to
radiation, and tumors that respond well to chemotherapy
60
would be controlled with radiation therapy even if they
had not been treated with chemotherapy. Another possibil-
ity is that chemotherapy induces or selects cell clones that
are resistant to radiation; however, this explanation is not
50
Miso supported by the fact that most of the drugs that have been
HBO
used for neoadjuvant chemotherapy do not produce cross-
HBO
resistant effects with radiation. The most likely explana-
tion seems to be that chemotherapy induces accelerated
40
repopulation of tumor cell clonogens, an explanation for
10 20 30 40 50 60
which some experimental evidence has been found.39
Treatment duration (days)
Inhibition of Tumor Angiogenesis
FIGURE 4-10. Values for the radiation dose yielding local tumor
The antitumor effects of chemotherapeutic agents are
control in 50% of the animals tested (TCD50) are plotted as a
function of overall treatment times for squamous cell carcinomas not restricted to killing tumor cells. They also include
of the head and neck. Total doses were normalized to the dose the ability to affect normal tissue structures within the
equivalent of that of a regimen of 2-Gy fractions using an Ä…/²
tumor, primarily by their cytotoxicity to endothelial cells
value of 25 Gy. Doses and times are best estimates of median
of tumor blood vessels. Angiogenesis (i.e., formation of
values. The dose and control rate reported in the literature from
blood vessels) is a prerequisite to the progressive growth
which the TCD50 values were calculated are shown as blue circles
of malignant tumors beyond small microscopic aggregates
to indicate the extent of the extrapolation. The rate of increase
of tumor cells. Endothelial cells in tumor blood vessels
in TCD50 was predicted from a 2-month clonogen-doubling rate
exhibit high rates of proliferation, making them sensitive
(dashed line). Estimated increases in TCD50 (solid lines) over the
to the cytotoxic actions of chemotherapeutic agents.
radiation-treatment period are shown for T2 tumors (purple cir-
Increasing evidence shows that chemotherapeutic agents
cles), T3 tumors (red squares), and tumors of mixed T stage
(green triangles). HBO, hyperbaric oxygen; Miso, misonidazole. have antiangiogenic effects.43-48 For example, docetaxel
(From Withers HR, Taylor JMG, Maciejewski B. The hazard of was reported to suppress endothelial cell proliferation and
accelerated tumor clonogen repopulation during radiotherapy.
tubule assembly in vitro and vessel formation in an in vivo
Acta Oncol 1988;27:131-146.)
Matrigel plug assay.43 These antiangiogenic effects were
partly blocked by the presence of the angiogenic factors
it also reduces the tumor-control efficacy of irradiation. vascular endothelial cell growth factor (VEGF) and basic
Any approach that reduces or eliminates accelerated clono- fibroblast growth factor (bFGF). This protective effect,
gen proliferation in tumors would improve the efficacy of however, could be overcome by anti-VEGF antibody or
radiation therapy. This is likely one of the major mechanisms by combining docetaxel with 2-methoxyestradiol, another
by which chemotherapeutic drugs improve local tumor con- antiangiogenic agent. In another study, docetaxel was
trol when given concurrently with radiation therapy. Even a shown to inhibit the in vivo formation of blood vessels in
small decrease in repopulation between radiation fractions mice at the intradermal injection site of tumor cells.44 This
can significantly improve tumor response to fractionated ra- antiangiogenic activity is probably another mechanism
diation therapy. However, care must be taken to select drugs by which taxanes (and probably other chemotherapeutic
that preferentially affect rapidly proliferating cells and pref- agents) enhance tumor response to irradiation, as sug-
erentially localize in malignant tumors. Another possibility gested by increasing preclinical evidence that inhibitors
is to supplement chemoradiation therapy with agents that of angiogenesis can enhance tumor radioresponse by a
have antiproliferative properties, such as those that block variety of mechanisms, including inhibition of VEGF and
membrane receptors for growth factors or interfere with bFGF, which typically act as radioprotectants for tumor
signaling pathways involved in cell proliferation.42 cells.49-51 VEGF is also a potent vessel permeability factor
Increased repopulation of tumor cells may also play an that causes fluid accumulation in extracapillary spaces and
important role in induction (neoadjuvant) chemotherapy. consequent impairment of blood flow and oxygen supply
The rationale for this form of therapy is to reduce the to tumor cells; inhibition of VEGF by chemotherapeutic
number of clonogenic cells, improve tumor oxygenation, agents would result in increased tumor oxygenation.
and eradicate tumor cells that are intrinsically resistant to Other preclinical evidence suggests that frequent ad-
radiation. According to radiobiologic principles, all of these ministration of small doses of chemotherapeutic agents
effects should improve tumor response to irradiation. Many preferentially targets tumor endothelial cells and that this
clinical trials have been conducted in which induction che- strategy has better antitumor efficacy and lower toxicity
motherapy has been combined with radiation therapy, but than high-dose chemotherapy given less frequently.45-48
the results have not been greatly encouraging. Although It is reasonable to expect that this approach, referred to
large proportions of patients in these trials responded as antiangiogenic chemotherapy, low-dose chemotherapy,
to chemotherapy by showing complete or partial tumor or metronomic chemotherapy, would enhance the effect
regression, significant improvement in treatment outcome of radiation therapy, perhaps even more so than high-dose
has yet to be achieved. Several possible explanations exist chemotherapy, in vivo.
Tumor dose (Gy) normalized to 2-Gy fractions
4 Principles of Combining Radiation Therapy and Chemotherapy 111
continued over the next few weeks. Alternating schedules
SEQUENCING OF CHEMOTHERAPY
such as these are given in an attempt to minimize exces-
AND RADIATION THERAPY
sive toxic effects on normal tissues (e.g., mucositis) and
Depending on the principal aim of chemoradiation ther- to enhance tumor response by exploiting some biologic
apy, drugs can be given before, during, or after the course effect of an agent that renders tumor cells more respon-
of radiation therapy. According to this sequencing, chemo- sive to the subsequent agent (e.g., perturbing cell cycling,
therapy is designated induction (neoadjuvant) chemo- reoxygenation).
therapy when it is given before irradiation, concurrent
Adjuvant Chemotherapy
chemotherapy when it is given during radiation therapy, and
adjuvant chemotherapy when it is given after irradiation. Adjuvant chemotherapy is given after a course of radiation
therapy is completed. The primary objective of adjuvant
Induction Chemotherapy
chemotherapy is to eradicate disseminated disease, although
Induction or neoadjuvant chemotherapy has two main drugs also can eradicate surviving tumor cells in irradiated
objectives. One is to eradicate micrometastases while they tumors.
contain small numbers of tumor cells. For this reason,
induction chemotherapy is begun soon after tumor diagnosis
CLINICAL APPLICATIONS OF
and can precede radiation therapy by weeks or months. The
other objective of induction chemotherapy is to reduce the
COMBINED RADIATION THERAPY
size of the primary tumor that is to be irradiated. Reducing
AND CHEMOTHERAPY
the number of clonogenic cells in the tumor in this way
increases the probability of tumor control by irradiation. Laboratory investigations are providing important pre-
Chemotherapy-induced tumor shrinkage may allow a clinical data regarding the interactions between combin-
smaller tissue volume to be irradiated, limiting damage to ations of cytotoxic drugs and radiation, but sequential
normal tissues. The latter strategy is particularly important clinical investigations are required to establish the toxicity,
in treating lymphomas and tumors in children. effectiveness, and value of combined chemotherapy and
Induction chemotherapy generally has not provided the radiation therapy for the care of human patients. The
therapeutic benefits that might be expected from theo- effects of drugs, given as single agents or in combination
retical principles. One reason is that chemotherapy, like with other drugs, have usually been investigated without
irradiation, may accelerate the proliferation of tumor cell the addition of radiation therapy. However, for diseases for
clonogens. Another possibility is that induction chemo- which radiation therapy is the standard of care, the addition
therapy may select for or induce drug-resistant cells that of chemotherapy before irradiation is common, even when
may be cross-resistant to radiation therapy. Development the value of such an addition is unknown. Separating the
of drug resistance remains a significant problem in chemo- administration of the drug and the radiation by 2 weeks or
therapy; nevertheless, evidence is lacking to convincingly more usually does not increase toxic reactions, and induction
demonstrate that cells that acquire drug resistance are also chemotherapy often produces some degree of response. The
resistant to radiation. Chemotherapeutic drugs can, under situation is quite different for concurrent chemotherapy and
certain conditions, increase the metastatic potential of radiation therapy, for which phase I clinical trials are needed
tumor cells or damage normal tissues in such a way as to to test a standard dose of radiation with increasing doses
make them conducive to metastatic growth. of chemotherapy or a standard dose of chemotherapy with
increasing doses of radiation. After the maximum tolerated
Concurrent Chemotherapy
combination of chemotherapy and radiation therapy is
In this mode of chemoradiation therapy, chemotherapeutic established, phase II trials can be used to test the efficacy of
drugs are given during a course of radiation treatment so the combination in one or more diseases.
that the drug or drugs exert their effects during several If phase II studies suggest that a particular combina-
radiation fractions. Even though drugs can act on systemic tion is efficacious, phase III comparative trials are required
and primary lesions, the main objective in concurrent to compare the standard treatment against the new com-
chemotherapy is to use drug-radiation interactions to bination. Rarely or never are phase II results sufficiently
maximize tumor radioresponse. The optimal schedule compelling to change practice throughout a nation or the
for drug administration (e.g., once each week, daily, at world. Comparisons with historical control subjects are
some other interval) depends on many factors, primarily often undertaken, but known or unsuspected biases often
the mechanisms of radioenhancement and the particular influence the interpretation of such comparisons.
drug and the conditions under which the enhancement is
highest and the toxicity of the treatment to normal tissue.
In this type of chemoradiation therapy, therapeutic benefit SEQUENTIAL CHEMOTHERAPY
occurs only when the enhancement of tumor response is
AND RADIATION THERAPY
greater than the toxic effects on critical normal tissues.
Induction Chemotherapy
Concurrent chemoradiotherapy has provided the best
clinical results in terms of local tumor control and patient Induction chemotherapy relies heavily on the principle of
survival. spatial cooperation. For epithelial tumors, little evidence
In one mode of concurrent chemoradiation therapy, exists to suggest that the addition of chemotherapy before
chemotherapy and radiation therapy are alternated rapidly irradiation confers any benefit in controlling locoregional
(e.g., radiation is given one week and chemotherapy dur- tumors; rather, the chemotherapy is directed at subclinical
ing the subsequent week, or vice versa), and this pattern is metastases. Only for malignant lymphomas does induction
112 PART 1 Principles
chemotherapy seem to be beneficial in terms of local tumor
Concurrent Chemotherapy and Radiation
control. For example, for diffuse large cell lymphomas that
Therapy
are less than 3.5 cm in their greatest dimension, induction
chemotherapy that produces a complete response coupled The concurrent use of chemotherapy and radiation ther-
with subsequent irradiation can lead to successful local apy tends to maximize the antitumor effect, even though it
control at a low total dose of radiation (e.g., 30 Gy). almost inevitably increases the acute toxicity of the treatment.
However, tumors 10 cm in diameter or larger that respond Concurrent chemoradiation therapy has been shown to
completely to induction chemotherapy nonetheless require improve survival of patients with several types of carcinoma
substantially higher total doses (at least 45 Gy) than do (Table 4-3). In a study of carcinoma of the esophagus,
smaller tumors.52 cisplatin and 5-fluorouracil (5-FU) given concurrently with
In contrast to lymphomas, locoregional control of carci- modest-dose irradiation (50 Gy) produced improved survival
nomas is not improved by the addition of induction che- relative to high-dose radiation therapy (64 Gy) alone.62 In
motherapy.53,54 The response of locoregional tumors to another study, the use of cisplatin concurrent with radiation
induction chemotherapy can be misleading. Many inves- therapy followed by cisplatin and 5-FU improved survival
tigators have been convinced that initial shrinkage of the over radiation therapy alone for patients with carcinoma of
tumor must eventually lead to improvements in locore- the nasopharynx.63 Concurrent chemotherapy and radiation
gional control. However, this has not been borne out for therapy also improved survival among patients with moderately
carcinomas of the upper aerodigestive tract54; moreover, advanced carcinoma of the cervix compared with radiation
phase III comparative trials of induction chemotherapy for therapy alone.64 Other supporting studies clearly confirmed
cancer of the cervix have shown survival to be better after the value of the combined treatment.65
radiation therapy alone than after induction chemotherapy Although the findings for carcinoma of the nasopharynx
followed by pelvic irradiation.55,56 are compelling, the role of concurrent chemotherapy and
In some circumstances, the use of induction chemo- radiation therapy for other sites in the upper aerodigestive
therapy before irradiation can improve survival over that tract is less clear. In one meta-analysis of head and neck
produced by radiation therapy alone. The best example is squamous cell carcinoma,54 an advantage was found for
non small cell carcinoma of the lung, for which cisplatin- concurrent chemotherapy and radiation therapy but not
based induction chemotherapy has been shown to improve for induction chemotherapy. In a trial comparing concur-
survival rates over those produced by radiation therapy rent chemoradiation with hyperfractionated irradiation to a
alone in three independent studies.57-59 Despite the fact total dose of 75 Gy for patients with locally advanced head
that induction chemotherapy has not been particularly and neck cancer,66 concurrent chemotherapy and radia-
effective in other diseases, many oncologists feel compelled tion therapy proved superior to irradiation alone in terms
to use chemotherapy before radiation. The rationales for of local control and might have been beneficial for overall
this practice include the convenience of being able to start survival.
chemotherapy quickly rather than having to wait to plan The improvement in survival produced by induction
radiation therapy, the ability to give the maximum doses chemotherapy for unresectable non small cell carcinoma
of chemotherapy if no radiation therapy is to be given, of the lung is based entirely on control of distant
and the long-standing belief that the subclinical burden metastasis. Two studies67,68 have tested the hypothesis that
is smallest at the time of diagnosis. However, because
induction chemotherapy does not affect local control
TABLE 4-3. Chemotherapy and Radiation Therapy
of epithelial tumors and because in some circumstances
for Carcinomas: Survival According to
(e.g., carcinoma of the cervix) starting chemotherapy
Therapy Sequence*
can delay effective local treatment and produce inferior
results, induction chemotherapy should be used only
Cancer Type Induction Concurrent Adjuvant
when a phase III comparative trial has demonstrated a
Head and neck54 Ä… +
clear benefit.
Nasopharynx63 +
The toxicity associated with combining chemotherapy
and radiation therapy can be minimized by separating them
NSCLC53,54,59,67,68 + +
in time, another feature that can make induction chemo-
SCLC54,69,79 + +
therapy more attractive. Occasionally, induction chemo-
therapy can even increase the protection of normal tissues.
Esophagus62,76,77 Ä…S +
This is true for patients with Hodgkin disease who present
Stomach80 S+
with a large mediastinal mass; shrinking the mass with che-
Rectum72-74 S+
motherapy means that more lung can be protected in the
course of subsequent radiation therapy.
Prostate60 +(H)
Adjuvant Chemotherapy Cervix64,65 - +
With one exception, adjuvant chemotherapy (i.e., chemo- Breast81,82 S+
therapy that is delivered after radiation therapy) is not
*
Based only on phase III comparative trials.
particularly valuable. That exception is in the use of
H, hormone therapy; NSCLC, non small cell lung cancer; SCLC,
adjuvant hormonal therapy for prostate cancer. Disease-
small cell lung cancer; S, surgical adjuvant radiation and chemo-
free and overall survival rates have been improved by
therapy; +, increased survival compared with radiation therapy; -,
following radiation therapy to the prostate with 2 to 3 years
decreased survival compared with radiation therapy; Ä…, conflicting
of androgen-suppression therapy.60,61 results (increased, decreased, and no difference).
4 Principles of Combining Radiation Therapy and Chemotherapy 113
concurrent chemoradiation would be superior to induc- had had only surgery. In another study of rectal cancer,72
tion chemotherapy by virtue of improving control of local postoperative chemotherapy delivered before, during,
disease and distant metastasis. In one, a comparative study,67 and after pelvic radiation therapy produced significantly
concurrent chemoradiation produced a statistically signifi- better survival rates than did postoperative pelvic radiation
cant improvement in overall survival over that produced therapy alone. The chemotherapy in that study included
by induction chemotherapy. Preliminary results from the bolus-dose 5-FU; results from a later study indicated that
Radiation Therapy Oncology Group (RTOG) protocol 9410 protracted-infusion 5-FU produced better results than did
also support the benefit of concurrent chemoradiation over bolus dosing.73 Another study by the National Surgical
induction chemotherapy followed by standard irradiation.68 Adjuvant Breast and Bowel Project group found that among
The value of radiation therapy in combination with che- patients with rectal cancer, survival after postoperative
motherapy for small cell carcinoma of the lung is unclear; chemoradiation was no better than that after postoperative
this type of cancer has been so responsive to combination chemotherapy alone.74 Controversies about the relative
chemotherapy that radiation therapy is considered by some contributions of chemotherapy and radiation therapy for
to be unnecessary. Several trials comparing chemotherapy rectal cancer are not likely to be resolved any time soon, as
alone versus chemotherapy with concurrent or adjuvant new chemotherapeutic agents are tested in postoperative
irradiation for small cell lung cancer have provided con- settings with or without pelvic irradiation and as more
flicting results. A meta-analysis suggested a small but sta- interest develops in the preoperative administration of
tistically significant benefit to the use of radiation therapy various combined treatments. Issues regarding the com-
over the use of chemotherapy alone. However, the optimal bination of radiation therapy with surgery are discussed in
timing and sequencing of the treatments remain controver- further detail in Chapter 3.
sial. A study at the National Cancer Institute of Canada69 Combinations of surgery, chemotherapy, and radiation
indicated that early use of chemoradiation therapy was therapy have been studied for potentially resectable carci-
superior to delayed irradiation of the thorax. The best noma of the esophagus. No prospective randomized trial
results found with regard to limited-stage small cell car- has shown that postoperative radiation therapy produces
cinoma of the lung have come from a national intergroup better results than surgical resection alone for carcinoma of
trial comparing accelerated fractionation with standard the esophagus, nor has any evidence been found to suggest
fractionation, both of which were given concurrently with that preoperative irradiation or preoperative chemother-
chemotherapy; however, that study did not address the apy is beneficial in this disease.75 However, considerable
difference between induction and concurrent chemoradia- interest has been expressed with regard to induction che-
tion. The only trial to have directly addressed that question motherapy given concurrently with radiation therapy fol-
indicated that the concurrent approach was superior.70 lowed by surgical resection. In one such study involving
In contrast to the diseases previously described, carci- adenocarcinoma of the esophagus that included tumors
noma of the cervix has a relatively favorable prognosis after of the gastric cardia, induction chemoradiation followed
radiation therapy alone. The 5-year survival rate for patients by surgery significantly improved survival rates compared
with invasive cancer of the cervix (all stages combined) with patients who underwent only surgery.76 However, the
treated with radiation therapy is approximately 60%. In- survival rates for the surgery-only group in that trial were
duction chemotherapy does not improve this percentage substantially lower than those reported in a later study.75 A
and can have a deleterious effect on survival. However, a larger study of chemoradiation given before surgical resec-
series of trials provided convincing evidence that concur- tion for squamous cell carcinoma of the esophagus showed
rent chemoradiation therapy can improve on the results no evidence of benefit from this treatment.77
of radiation therapy. The RTOG protocol 9001 compared Tests of concurrent chemoradiation in the postoperative
concurrent cisplatin and 5-FU with pelvic and intracavi- setting have shown evidence of benefit. A phase III com-
tary irradiation versus radiation therapy alone for cervical parative trial of concurrent cisplatin and external radiation
cancer.64 The concurrent chemoradiation strategy pro- therapy versus external radiation therapy alone was con-
duced significantly better disease-free and overall survival ducted in patients with high-risk squamous cell carcinoma
rates and provided better local tumor control and freedom of the head and neck after total resection of all evident
from distant metastasis. Other cooperative trials from the disease. The rate of locoregional recurrence was signifi-
Gynecologic Oncology Group and the Southwest Oncol- cantly lower in the concurrent therapy group than in the
ogy Group have emphasized the importance of concurrent radiation-only group. However, survival rates were no dif-
cisplatin in treating cervical cancer.65 ferent between the two groups, and the incidence of severe
acute adverse effects in the combined-modality group was
Chemoradiation plus Surgery
more than double that in the irradiation-only group.78
Chemotherapy and radiation have been used in combination In summary, the sequence of combination treatment that
with surgery in attempts to produce better survival rates has produced the best results seems to be concurrent chemo-
and organ conservation than can be obtained by surgical radiation therapy (see Table 4-3).53,54,59-65,67-69,72-74,76,77,79-82
treatment alone. However, the results from trimodality However, considerable variation is evident among disease
treatments are difficult to compare and interpret by virtue sites and according to whether surgery was also used in the
of the complexities of treatment sequencing and timing, and treatment.
the relative contributions of the modalities remain unclear.
Organ Conservation
For example, the Gastrointestinal Tumor Study Group71
reported that use of combined chemoradiation therapy Conservation of organ structure and function remains an
after surgery prolonged disease-free survival time among important goal even when survival may not be improved
patients with rectal carcinoma compared with those who by treatment. One of the earliest demonstrations of this
114 PART 1 Principles
principle was the use of radiation therapy to avoid the laryngectomy followed by radiation therapy or to receive
need for a permanent colostomy in patients with advanced neoadjuvant therapy with cisplatin and 5-FU followed by
anal carcinoma. Chemotherapy combined with radiation radiation therapy for those who achieved a 50% or better
therapy is now considered the treatment of choice for tumor response to the chemotherapy. Survival among the
carcinoma of the anal canal.83 A phase III trial is being first group (i.e., surgery plus radiation) was slightly but not
conducted to address whether mitomycin can improve significantly better than among those who had laryngectomy
outcome compared with the use of 5-FU and irradiation alone, but laryngeal function could be preserved in more
for rectal cancer (RTOG protocol 8704).84 Although some than 60% of patients who underwent chemoradiation.
physicians prefer to use a combination of 5-FU and cisplatin These findings led to a large intergroup trial, coordinated
with radiation therapy, the equivalence or superiority by the RTOG, in which initial chemotherapy followed by
of these drugs to 5-FU and mitomycin has not been radiation therapy was compared with radiation therapy
demonstrated. alone and with a third arm in which radiation therapy
was given concurrently with cisplatin. The results of this
Rectal Carcinoma
trial showed that concurrent chemotherapy and radiation
The success of chemoradiation therapy in preserving therapy resulted in higher rates of locoregional control
function in patients with anal carcinoma led to widespread and laryngeal preservation than induction chemotherapy
interest in the use of preoperative chemoradiation for followed by radiation therapy or radiation therapy alone.86
carcinoma of the distal rectum (i.e., lesions occurring less
Bladder Carcinoma
than 6 cm from the anal verge). Traditional treatment
for these lesions involves abdominal peritoneal resection Given sufficient motivation on the part of both patients
and permanent colostomy. However, for some patients, and physicians, bladder conservation is possible in some
preoperative chemoradiation (usually involving 5-FU, cases of invasive bladder cancer through the use of
alone or in combination with other drugs) may reduce chemotherapy, radiation therapy, and surgery. The most
the tumor volume and peripheral extensions so that more common sequence is to begin with transurethral resection
limited surgical procedures that do not require a colostomy of the bladder tumor, followed by two or more cycles of
can be performed. No prospective studies have compared chemotherapy with methotrexate, cisplatin, and vinblastine,
chemoradiation followed by limited resection versus limited followed by pelvic irradiation with concomitant cisplatin.
resection alone versus abdominoperineal resection. Careful follow-up using cystoscopy and biopsy is required,
and cystectomy is reserved for persistent disease. In two
Laryngeal Carcinoma
studies of this approach, bladder conservation was possible
Total laryngectomy is accepted as being the single most for 40% to 50% of patients with T2 to T4a carcinomas of
effective treatment for laryngeal carcinoma. However, the bladder.87,88
radiation therapy can be used to treat T1 and T2 lesions
Toxicity Issues
while allowing conservation of the larynx and laryngeal
speech. Interest in laryngeal conservation for more All chemotherapeutic agents have clinically important
advanced lesions was generated by results of a clinical toxic effects. An overview of the organ-specific toxicity
trial within the U.S. Veterans Administration system.85 In of commonly used drugs or classes of drugs is provided
that trial, patients were randomized to receive standard in Table 4-4. Most chemotherapeutic agents are toxic to
TABLE 4-4. Effects of Cytotoxic Drugs on Normal Tissues
Drugs of Classes Bone Marrow GI Liver Kidney Lung Heart PNS/CNS Gonads
Anthracycines +++ + + +++
Alkylators +++ ++ ++ + + ++
Antifolates +++ ++ + ++ + +/+++
Antipyrimidines +++ +++ + + + +
Dacarbazine +++
Taxanes +++ + + +++/+
Podophyllotoxins ++ + +
Hydroxyurea ++ +
Mitomycin ++ +++ ++ +
Procarbazine ++ +
Vinca alkaloids + ++ +++/+
Cisplatin + +++ +++ ++/+
Bleomycin + +++
CNS, central nervous system; GI, gastrointestinal; PNS, peripheral nervous system; +, some reports of toxicity, usually not major; ++, major
toxicity not unusual; +++, principal toxicity expected.
4 Principles of Combining Radiation Therapy and Chemotherapy 115
11. Radford IR. Evidence for a general relationship between the
bone marrow; however, several also are toxic to other
induced level of DNA double-strand breakage and cell-killing
organs, which must be taken into account when combining
after x-irradiation of mammalian cells. Int J Radiat Biol 1986;49:
chemotherapy with radiation therapy.
611-620.
The sequential use of chemotherapeutic agents and
12. Kinsella TJ, Dobson PP, Mitchell JB, et al. Enhancement of x-ray
radiation therapy usually does not increase toxicity.
induced DNA damage by pretreatment with halogenated pyrimi-
Although both modalities affect the bone marrow, the
dine analogs. Int J Radiat Oncol Biol Phys 1987;13:733-739.
13. Wang Y, Pantelias GE, Iliakis G. Mechanism of radiosensitization
effects of radiation on bone tend to be localized unless
by halogenated pyrimidines: the contribution of excess DNA and
the irradiated volume is very large. The sequential use of
chromosome damage in BrdU radiosensitization may be minimal
bleomycin and radiation therapy is an important excep-
in plateau-phase cells. Int J Radiat Biol 1994;66:133-142.
tion, because the pulmonary toxicity of bleomycin can be
14. Elkind MM, Sutton H. X-ray damage and recovery in mammalian
greatly accentuated by subsequent radiation therapy. This
cells in culture. Nature 1959;184:1293-1295.
effect is probably of greatest importance in the treatment
15. Little JB. Repair of potentially-lethal radiation damage in mam-
of Hodgkin disease. malian cells: enhancement by conditioned medium from station-
ary cultures. Int J Radiat Biol 1971;20:87.
Concurrent chemoradiation therapy is associated with
16. Gregoire V, Hittelman WN. Nucleoside analogs as radiosensitiz-
much greater toxicity than is the sequential use of either
ing agents. In Cheson BD, Keating MJ, Plunkett W (eds): Nu-
modality. Drugs such as doxorubicin and gemcitabine,
cleoside Analogs in Cancer Therapy. New York: Marcel Dekker,
when given at the standard doses used to treat systemic
1997, 315-358.
disease, profoundly increase the effects of radiation ther-
17. Gregoire V, Hittelman WN, Rosier JF, et al. Chemo-radiotherapy:
apy. These combinations can lead to life-threatening toxic radiosensitizing nucleoside analogues. Oncol Rep 1999;6:949-957.
effects at total doses of radiation that are usually well toler- 18. Lawrence TS, Chang EY, Hahn TM, et al. Radiosensitization of
pancreatic cancer cells by 22 ,22 difluoro-22 -deoxycytidine. Int J
ated. Similarly, the concurrent use of chemotherapy and
Radiat Oncol Biol Phys 1996;34:867-872.
irradiation can greatly increase esophageal reactions when
19. Plunkett W, Huang P, Gandhi V. Metabolism and action of
cisplatin and etoposide are given concurrently with irradia-
fludarabine phosphate. Semin Oncol 1990;17:3-17.
tion for carcinoma of the lung.
20. Plunkett W, Huang P, Xu YZ, et al. Gemcitabine: metabolism,
Two major strategies are available to enhance the
mechanisms of actions, and self-potentiation. Semin Oncol
1995;22:3-10.
effectiveness of concurrent treatment while avoiding its
21. Lawrence TX, Eisbruch A, Shewach DS. Gemcitabine-mediated
toxic effects. The first strategy is to use conformal radia-
radiosensitization. Semin Oncol 1997;24:24-28.
tion therapy to specifically avoid irradiating normal tis-
22. Gregoire V, Hunter N, Milas L, et al. Potentiation of radiation-
sues that have the most severe reactions.89 The second
induced regrowth delay in murine tumors by fludarabine. Cancer
strategy is to use cytoprotective agents such as amifos-
Res 1994;54:468-474.
tine.90 These and other agents are being investigated for
23. Gregoire V, Hunter NR, Brock WA, et al. Improvement in the
the degree of protection they produce during concurrent therapeutic ratio of radiotherapy for a murine sarcoma by in-
domethacin plus fludarabine. Radiat Res 1996;146:548-553.
chemotherapy and radiation therapy. Newer cytoprotec-
24. Milas L, Fujii T, Hunter N, et al. Enhancement of tumor radiore-
tive drugs may become available within the next few years
sponse in vivo by gemcitabine. Cancer Res 1999;59:107-114.
that may enhance the therapeutic ratio of concurrent che-
25. Terasima T, Tolmach LJ. Variations in several responses of HeLa
motherapy and radiation therapy, which remains the most
cells to x-irradiation during the cell cycle. Biophys J 1963;3:11-33.
compelling chemoradiation strategy for achieving tumor
26. Schiff PB, Fant J, Horwitz SB. Promotion of microtubule assem-
control. bly in vitro by taxol. Nature 1979;277:665-667.
27. Tishler RB, Geard CR, Hall EJ, et al. Taxol sensitizes human astro-
cytoma cells to radiation. Cancer Res 1992;52:3495-3497.
REFERENCES
28. Milas L, Milas MM, Mason KA. Combination of taxanes with
radiation: preclinical studies. Semin Radiat Oncol 1999;9:12-26.
1. Steel GG, Peckham MJ. Exploitable mechanisms in combined
29. Gregoire V, Van NT, Brock WA, et al. The role of fludarabine-
radiotherapy-chemotherapy: the concept of additivity. Int J Radiat
induced apoptosis and cell cycle synchronization in enhanced
Oncol Biol Phys 1979;5:85-91.
murine tumor radiation response in vivo. Cancer Res 1994;54:
2. Yuhas JM, Storer JB. Differential chemo-protection of normal
6201-6209.
and malignant tissues. J Natl Cancer Inst 1969;42:331-335.
30. Milas L, Hunter NR, Mason KA, et al. Role of reoxygenation in
3. Milas L, Hanson WR. Eicosanoids and radiation. Eur J Cancer
induction of enhancement of tumor radioresponse by paclitaxel.
1995;31A:1580-1585.
Cancer Res 1995;55:3564-3568.
4. Hahn SM, Krishna MC, Samuni A, et al. Potential use of nitrox-
31. Brown JM, Giaccia AJ. The unique physiology of solid tumors:
ides in radiation oncology. Cancer Res 1994;54:2006S-2010S.
opportunities (and problems) for cancer therapy. Cancer Res
5. Buntzel J, Kuttner K, Frohlich D, et al. Selective cytoprotection
1998;58:1408-1416.
with amifostine in concurrent radiochemotherapy for head and
32. Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nu-
neck cancer. Ann Oncol 1998;9:505-509.
trient supply, and metabolic microenvironment of human tumors:
6. Hall EJ. Radiobiology for the Radiologist, 4th ed. Philadelphia: JB
a review. Cancer Res 1989;49:6449-6465.
Lippincott, 1993.
33. Tannock IF, Rotin D. Acid pH in tumors and its potential for
7. Kallman RF (ed). Rodent Tumor Models in Experimental Cancer
therapeutic exploitation. Cancer Res 1989;49:4373-4384.
Therapy. New York: Pergamon Press, 1987.
34. Brown JM, Biaglow JE, Hall EJ, et al. Sensitizers and protectors
8. Begg AC. Principles and practices of the tumor growth delay as-
to radiation and chemotherapeutic drugs. Cancer Treat Symp
say. In Kallman RF (ed): Rodent Tumor Models in Experimental
1984;1:85-101.
Cancer Therapy. New York: Pergamon Press, 1987, pp 114-121.
35. Hermens AF, Barendsen GW. The proliferative status and clono-
9. Suit HD, Sedlacek R, Thames HD Jr. Radiation dose-response
genic capacity of tumour cells in a transplantable rhabdomyosar-
assays of tumor control. In Kallman RF (ed): Rodent Tumor Mod-
coma of the rat before and after irradiation with 800 rad of x-rays.
els in Experimental Cancer Therapy. New York: Pergamon Press,
Cell Tissue Kinet 1978;11:83-100.
1987, pp 138-148.
36. Stephens TC, Steel GG. Regeneration of tumors after cytotoxic
10. Mason KA, Hunter NR, Milas M, et al. Docetaxel enhances tumor
treatment. In Meyn RE, Withers HR (eds): Radiation Biology in
radioresponse in vivo. Clin Cancer Res 1997;3:2431-2438.
Cancer Research. New York: Raven Press, 1980, pp 385 295.
116 PART 1 Principles
37. Trott KR. Cell repopulation and overall treatment time. Int J 57. Dillman RO, Seagren SL, Propert KJ, et al. A randomized trial
Radiat Oncol Biol Phys 1990;19:1071-1075. of induction chemotherapy plus high-dose radiation versus radia-
38. Milas L, Yamada S, Hunter N, et al. Changes in TCD50 as a tion alone in stage III non small-cell lung cancer. N Engl J Med
measure of clonogen doubling time in irradiated and unirradiated 1990;323:940-945.
tumors. Int J Radiat Oncol Biol Phys 1991;21:1195-1202.
58. Le Chevalier T, Arriagada R, Quoix E. Radiotherapy alone ver-
39. Milas L, Nakayama T, Hunter N, et al. Dynamics of tumor cell sus combined chemotherapy and radiotherapy in nonresectable
clonogen repopulation in a murine sarcoma treated with cyclo- non small cell lung cancer. First analysis of a randomized trial in
phosphamide. Radiother Oncol 1994;30:247-253. 353 patients. J Natl Cancer Inst 1991;83:417-423.
40. Withers HR, Taylor JMG, Maciejewski B. The hazard of accelerated 59. Sause WT, Scott C, Taylor S, et al. Radiation Therapy Oncology
tumor clonogen repopulation during radiotherapy. Acta Oncol Group (RTOG) 88-08 and Eastern Cooperative Oncology Group
1988;27:131-146. (ECOG) 4588: preliminary results of a phase III trial in regionally
41. Taylor JMG, Withers HR, Mendenhall WM. Dose-time consid- advanced, unresectable non small cell lung cancer. J Natl Cancer
erations of head and neck squamous cell carcinomas treated with Inst 1995;87:198-205.
irradiation. Radiother Oncol 1990;17:95-102.
60. Pilepich MV, Caplan R, Byhardt RW, et al. Phase III trial of
42. Mendelsohn J, Fan Z. Epidermal growth factor receptor family androgen suppression using goserelin in unfavorable-prognosis
and chemosensitization. J Natl Cancer Inst 1997;89:341-343. carcinoma of the prostate treated with definitive radiotherapy:
43. Sweeney CJ, Miller KD, Sissons SE, et al. The antiangiogenic report of Radiation Therapy Oncology Group Protocol 85-31.
property of docetaxel is synergistic with a recombinant human- J Clin Oncol 1997;15:1013-1021.
ized monoclonal antibody against vascular endothelial growth fac- 61. Bolla M, Gonzalez D, Warde P, et al. Improved survival in patients
tor or 2-methoxyestradiol but antagonized by endothelial growth with locally advanced prostate cancer treated with radiotherapy
factors. Cancer Res 2001;61:3369-3372. and goserelin. N Engl J Med 1997;337:295-300.
44. Schimming R, Hunter NR, Mason KA, et al. Apoptosis and in- 62. Herskovic A, Martz K, Al-Sarraf M, et al. Combined chemotherapy
hibition of neoangiogenesis as mechanisms of antitumor action and radiotherapy compared with radiotherapy alone in patients
by docetaxel (Taxotere). Mund Kiefer Gesichtschir 1999;3: with cancer of the esophagus. N Engl J Med 1992;326:1593-1598.
210-212.
63. Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy ver-
45. Browder T, Butterfield CE, Kräling BM, et al. Antiangiogenic sus radiotherapy in patients with advanced nasopharyngeal can-
scheduling of chemotherapy improves efficacy against experi- cer: phase III randomized Intergroup study 0099. J Clin Oncol
mental drug-resistant cancer. Cancer Res 2000;60:1878-1886. 1998;16:1310-1317.
46. Klement G, Baruchel S, Rak J, et al. Continuous low-dose ther- 64. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent
apy with vinblastine and VEGF receptor-2 antibody induces chemotherapy compared with pelvic and para-aortic radiation for
sustained tumor regression without overt toxicity. J Clin Invest high-risk cervical cancer. N Engl J Med 1999;340:1137-1143.
2000;105:R15-R24.
65. Thomas GM. Improved treatment for cervical cancer: concur-
47. Hanahan D, Bergers G, Bergsland E. Less is more, regularly: met- rent chemotherapy and radiotherapy. N Engl J Med 1999;340:
ronomic dosing of cytotoxic drugs can target tumor angiogenesis 1198-1200.
in mice. J Clin Invest 2000;105:1045-1047.
66. Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated
48. Hahnfeldt P, Folkman J, Hlatky L. Minimizing long-term tumor irradiation with or without concurrent chemotherapy for
burden: the logic for metronomic chemotherapeutic dosing and locally advanced head and neck cancer. N Engl J Med 1998;338:
its antiangiogenic basis. J Theor Biol 2003;220:545-554. 1798-1804.
49. Mason KA, Komaki R, Cox JD, et al. Biology-based combined- 67. Furuse K, Fukuoka M, Kawahara M, et al. Phase III study of con-
modality radiotherapy: workshop report. Int J Radiat Oncol Biol current versus sequential thoracic radiotherapy in combination
Phys 2001;50:1079-1089. with mitomycin, vindesine, and cisplatin in unresectable stage III
50. Gorski DH, Beckett MA, Jaskowiak NT, et al. Blockage of the non small-cell lung cancer. J Clin Oncol 1999;17:2692-2699.
vascular endothelial growth factor stress response increases the 68. Curran WJ, Scott C, Langer C, et al. Phase III comparison of
antitumor effects of ionizing radiation. Cancer Res 1999;59: sequential vs concurrent chemoradiation for PTS with unresected
3374-3378. stage III non small cell lung cancer (NSCLC): initial report of Ra-
51. Haimovitz-Friedman A, Vlodavsky I, Chaudhur A, et al. Autocrine diation Therapy Oncology Group (RTOG) 9410 [abstract]. Proc
effects of fibroblast growth factor in repair of radiation damage in Am Soc Clin Oncol 2000;19:484a.
endothelial cells. Cancer Res 1991;51:2552-2558. . Murray N, Coy P, Pater JL, et al. Importance of timing for tho-
52. Wilder RB, Tucker SL, Ha CS, et al. Dose-response analysis for racic irradiation in the combined modality treatment of limited-
radiotherapy delivered to patients with intermediate grade and stage small-cell lung cancer. J Clin Oncol 1993;11:336-344.
large-cell immunoblastic lymphomas that have completely re- 70. Goto K, Nishiwaki Y, Takada M, et al. Final results of a phase
sponded to CHOP-based induction chemotherapy. Int J Radiat III study of concurrent versus sequential thoracic radiotherapy
Oncol Biol Phys 2001;49:17-22. (TRT) in combination with cisplatin (P) and etoposide (E) for
53. Arriagada R, Le Chevalier T, Quoix E, et al. ASTRO (American limited-stage small cell lung cancer (LD-SCLC): the Japan Clini-
Society for Therapeutic Radiology and Oncology) plenary: effect cal Oncology Group (JCOG) study [abstract]. Proc Am Soc Clin
of chemotherapy on locally advanced non small cell lung car- Oncol 1999;18:468a.
cinoma: a randomized study of 353 patients. GETCB (Groupe 71. Gastrointestinal Tumor Study Group. Prolongation of the disease-
d Etude et Traitement des Cancers Bronchiques), FNCLCC free interval in surgically treated rectal carcinoma. N Eng J Med
(Féderation Nationale des Centres de Lutte contre le Cancer) 1985;312:1465-1472.
and the CEBI trialists. Int J Radiat Oncol Biol Phys 1991;20: 72. Krook JE, Moertel CG, Gunderson LL, et al. Effective surgical
1183-1190. adjuvant therapy for high-risk rectal carcinoma. N Engl J Med
54. Pignon JP, Bourhis J, Domenge C, et al. Chemotherapy added 1991;324:709-715.
to locoregional treatment for head and neck squamous-cell car- 73. O Connell MJ, Martenson JA, Wieand HS, et al. Improving
cinoma: three meta-analyses of updated individual data. Lancet adjuvant therapy for rectal cancer by combining protracted-
2000;355:949-955. infusion fluorouracil with radiation therapy after curative surgery.
55. Tattersall MH, Lorvidhaya V, Vootiprux V, et al. Randomized N Engl J Med 1994;331:502-507.
trial of epirubicin and cisplatin chemotherapy followed by pel- 74. Wolmark N, Wieand HS, Hyams DM, et al. Randomized trial
vic radiation in locally advanced cervical cancer. J Clin Oncol of postoperative adjuvant chemotherapy with or without radio-
1995;13:444-451. therapy for carcinoma of the rectum: National Surgical Adju-
56. Souhami L, Gil RA, Allan SE, et al. A randomized trial of vant Breast and Bowel Project protocol R-02. J Natl Cancer Inst
chemotherapy followed by pelvic radiation therapy in stage IIIB 2000;92:388-396.
carcinoma of the cervix. J Clin Oncol 1991;9:970-977.
4 Principles of Combining Radiation Therapy and Chemotherapy 117
75. Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed III randomized trial of the European Organization for Research
by surgery compared with surgery alone for localized esophageal and Treatment of Cancer Radiotherapy Gastrointestinal Coopera-
cancer. N Engl J Med 1998;339:1979-1984. tive Groups. J Clin Oncol 1997;15:2040-2049.
76. Walsh TN, Noonan N, Hollywood D, et al. A comparison of 84. Flam MS, Pajak TF, Haller D, et al. The role of mitomycin in com-
multimodal therapy and surgery for esophageal adenocarcinoma. bination with 5-fluorouracil and radiotherapy in the definitive
N Engl J Med 1996;335:462-467. nonsurgical treatment of epidermoid carcinoma of the anal canal:
77. Bosset J-F., Gignou M, Triboulet J-P, et al. Chemoradiotherapy 5-year updated results of a phase III randomized study (RTOG
followed by surgery compared with surgery alone in squamous- 8704/ECOG 1289) [abstract]. Presented at the Japan/US Cancer
cell cancer of the esophagus. N Engl J Med 1997;337:161-167. Therapy Symposium, Hiroshima, Japan, April 1999.
78. Cooper JS, Pajak TF, Forastiere AA, et al. Postoperative concurrent 85. Spaulding MB, Fischer SG, Wolf GT. Tumor response, toxicity,
radiotherapy and chemotherapy for high-risk squamous-cell carci- and survival after neoadjuvant organ-preserving chemotherapy
noma of the head and neck. N Engl J Med 2004;350:1937-1944. for advanced laryngeal carcinoma. The Department of Veterans
79. Turrisi A, Kim K, Sause W, et al. Observations after 5 year follow- Affairs Cooperative Laryngeal Cancer Study Group. J Clin Oncol
up of intergroup trial 0096: 4 cycles of cisplatin (P) etoposide (E) 1994;12:1592-1599.
and concurrent 45 Gy thoracic radiotherapy (TRT) given in daily 86. Forastiere AA, Goepfert H, Maor M, et al. Concurrent chemo-
(QD) or twice-daily fractions followed by 25 Gy PCI. Survival therapy and radiotherapy for organ preservation in advanced la-
differences and patterns of failure [abstract]. Proc Am Soc Clin ryngeal cancer. N Engl J Med 2003;349:2091-2098.
Oncol 1998;17:457a.
87. Kaufman DS, Shipley WU, Griffin PP, et al. Selective bladder
80. Macdonald JS, Smalley S, Benedetti J, et al. Postoperative com- preservation by combination treatment of invasive bladder cancer.
bined radiation and chemotherapy improves disease-free survival N Engl J Med 1993;329:1377-1382.
(DFS) and overall survival (OS) in resected adenocarcinoma of the 88. Tester W, Caplan R, Heaney J, et al. Neoadjuvant combined mo-
stomach and GE junction. Results of intergroup study INT-0116 dality program with selective organ preservation for invasive blad-
(SWOG 9008) [abstract]. Proc Am Soc Clin Oncol 2000;19:1a. der cancer: results of Radiation Therapy Oncology Group phase II
81. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radio- trial 8802. J Clin Oncol 1996;14:119-126.
therapy in high-risk premenopausal women with breast cancer 89. Fossella FV, Zinner RG, Komaki R, et al. Gemcitabine (G)
who receive adjuvant chemotherapy. N Engl J Med 1997;337: with concurrent chest radiation (XRT) followed by consolida-
949-955. tion chemotherapy with gemcitabine plus cisplatin (CDDP): a
82. Ragaz J, Jackson SM, Le N, et al. Adjuvant radiotherapy and phase I trial for patients with stage III non small cell lung cancer
chemotherapy in node positive premenopausal women with (NSCLC) [abstract]. J Clin Oncol 2001;20:312a.
breast cancer. N Engl J Med 1997;337:956-962.
90. Komaki R, Lee JS, Milas L, et al. Effects of amifostine on acute
83. Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radio- toxicity from concurrent chemotherapy and radiotherapy for
therapy and chemotherapy is superior to radiotherapy alone in inoperable non small cell lung cancer: report of a randomized com-
the treatment of locally advanced anal cancer: results of a phase parative trial. Int J Radiat Oncol Biol Phys 2004;58:1369-1377.