Paclitaxel combined with fractionated radiation in vitro: A study with vulvar squamous cell carcinoma cell lines

Abstract

Concurrent paclitaxel and radiation has given promising results in the treatment of a variety of solid tumors. We wanted to test the efficacy of this combination for vulvar carcinoma, which currently has a poor outcome in advanced stages. The radiation sensitivity, sublethal damage repair (SLDR) capacity and effect of paclitaxel during fractionated radiation were assessed in our study on 7 vulvar inherently radioresistant squamous cell carcinoma (SCC) cell lines. The 96-well plate clonogenic assay was used. Survival data were fitted to the linear quadratic model. The area under the curve (AUC), equivalent to mean inactivation dose (D̄), was obtained with numerical integration. AUC ratios between single-dose radiation and fractionated radiation with or without paclitaxel were used to determine the SLDR of the cell lines and the effect of paclitaxel on it. Seven currently tested vulvar SCC cell lines were found to have a limited capacity of repairing sublethal damage (SLD). Only 3 of them presented SLDR of significance. The effect of concurrent radiation and paclitaxel was clearly additive when the radiation dose was fractionated in most of the cell lines. In addition, 2 of the cell lines having SLDR exhibited a trend toward losing the repair capacity when paclitaxel was present during the irradiation. In addition, the survival curve of the UM-SCV-1A cell line gave the impression of a true paclitaxel effect on SLDR. Paclitaxel used concurrently with fractionated radiation showed effectiveness on vulvar carcinoma. The effect was at least additive and could even be expected to abrogate the SLDR during split-dose radiation. © 2001 Wiley-Liss, Inc.

Paclitaxel is a antimicrotubule anticancer agent with considerable efficacy in several human tumors.1 The mechanism of action of paclitaxel is based on promotion of microtubule assembly and the stabilization of these microtubules. This results in arrest in the G₂/M phase of the cell cycle.2 The G₂/M phase is known to be the most radiosensitive phase in the cell cycle and consequently, paclitaxel has aroused wide interest for its ability to act as a radiosensitizer in the treatment of malignant tumors. In vitro studies of concurrent paclitaxel and radiation have not yet led to congruent opinion. The results obtained with this combined treatment modality in vitro seem to be somewhat less than has been expected. Both supraadditive3–6 and additive7–10 results have been reported in a variety of malignant cell lines, but subadditive interaction has also been demonstrated in vitro.11, 12 Contradictory outcomes in cytotoxic studies on the mechanism of action of paclitaxel have thrown doubt on the accumulation in G₂/M being the main reason for the antitumor activity of paclitaxel.

The evidence has led to the conclusion that there are multiple pathways leading to cell death after paclitaxel exposure. The role of genes, for example, p53 and bcl-2, in regulating the cytotoxic effects of anticancer agents13, 14 is being examined. However, radiosensitizing effects demonstrated in preclinical models with some malignant tumors have encouraged clinicians to test combined therapy with paclitaxel and radiation in advanced cases of nonsmall cell lung, head and neck and breast cancer, among others, to determine the role of this therapy as a radiosensitizer.15, 16.

Most (90%) vulvar malignancies are squamous cell carcinomas (SCCs). The prognosis in advanced cases (stages III and IV) is poor and 5-year survival rates are 48% and 15%, respectively.17 Surgery is currently the mainstay of treatment and radiotherapy is included when lymph nodes are involved. Radiation alone has not been widely used in the management of vulvar cancer because of technical difficulties in directing the beam and also because of normal tissue reactions. Exceptional inherent radioresistance in vitro has been found in vulvar SCC,18, 19 and chemotherapy alone has not proved sufficient in the treatment of vulvar carcinoma. There are some reports of simultaneous radiotherapy and chemotherapy, using platinum analog-based combinations, that have shown promising results in treating patients with advanced or recurrent vulvar carcinoma as an alternative to radical surgery.20–23 Many vulvar carcinoma patients are at advanced ages, when extensive surgery is often unsatisfactory and therefore other treatment models are welcome.

We have recently tested the effects of concomitant, single-dose paclitaxel and radiation on vulvar SCC cell lines and found a clear additive interaction between these 2 modalities. Because in clinical radiotherapy the radiation dose is fractionated, we wanted to test the effects of paclitaxel concurrently with fractionated radiation on vulvar SCC cell lines and also determine the sublethal damage repair (SLDR) capacity in these lines and the effect of paclitaxel on it.

MATERIAL AND METHODS

Cell lines and cell culture

The cell lines used in our study were established from SCCs of the vulva at the University of Michigan (UM), Ann Arbor, MI and the University of Turku (UT), Turku, Finland.24, 25 The long-established vulvar SCC cell line A-431 was obtained from the American Type Culture Collection (Rockville, MD).26 Some characteristics of the cell lines are listed in Table I. Prior to the experiments, the cell lines were grown in DMEM containing 2 mM glutamine, 1% nonessential amino acids, 100 U/ml penicillin, 100 U/ml streptomycin and 10% FBS. The cells were kept in logarithmic growth by weekly or biweekly passage.

Table I. Characteristics of the Vulvar SCC Cell Lines1

Cell line	Age (yr)	TNM classification	SCC grade	Prior therapy	Specimen site	Doubling time in vitro (hr)
1 SCC, squamous cell carcinoma; nk, not known.
UM-SCV-1A	62	T3N2M1	Well-poor	None	Primary	32
UM-SCV-2	86	T3N1M0	Poor	Surgery	Local recurrence	58
UM-SCV-4	41	T2N2M0	Well	None	Primary	56
UM-SCV-6	43	T1N1M0	Mod.	Surgery	Primary	53
UM-SCV-7	77	T2N2M0	Well-poor	None	Primary	37
UT-SCV-3	73	T3N2M0	Mod.	None	Primary	13
A431	nk	nk	SCC	nk	Primary	14

Drug preparation

Paclitaxel (Taxol®, Bristol-Myers Squibb, Espoo, Finland) was received as an infusion concentrate of 6 mg/ml. A stock solution of 100 nM was prepared in Ham's F-12 medium, kept at −18°C and thawed immediately before the experiments. We had previously tested the paclitaxel-sensitivities of vulvar SCC cell lines;10, 19 for the current study, we chose paclitaxel doses expected to cause 20% inhibition in clonogenic survival. Final dilutions of 0.45–1.2 nM were used in the experiments.

Clonogenic assay and irradiation

The cells were grown in T25 culture flasks into midlogarithmic phase (40–60% confluency) and fed with fresh medium on the day before plating for the experiments. The clonogenic assay was performed as described previously.27, 28 Briefly, the cells were harvested with trypsin and EDTA, counted and diluted to a stock solution of 4,167 cells/ml. The number of cells plated per well was adjusted according to the plating efficiency (PE) of each cell line. Further dilutions of this single cell solution either with or without paclitaxel were made in 50 ml of Ham's F-12 medium containing 15% FBS and desired concentrations of paclitaxel were added to these dilutions. The cells were plated in 96-well culture plates by applying 200 μl/well using an octapipette. After plating, the cells were allowed to attach for 24 hr prior to irradiation. To test the concomitant use of paclitaxel and radiation, the cells were treated with paclitaxel for 24 hr before irradiation and the drug was allowed to remain in the plates during the whole incubation period. Two to 6 experiments including duplicate plates were carried out on each cell line to test the effect of concomitant paclitaxel and radiation at a single dose and in 2 fractions. Control plates without paclitaxel were irradiated simultaneously. The same single-cell solution was always used as the source of cells in 1 experiment.

The cells were irradiated in plates with 4 MeV photons generated by a linear accelerator (Clinac 4/100, Varian, Palo Alto, CA), which delivers a dose rate of 2 Gy/min. The cells were treated as follows: 0, 1.25, 2.5, 5 and 7.5 Gy either in a single dose or divided in 2 equal fractions with a 24 hr interval. This interval was chosen to give the cells enough time to carry out their repair processes and because it is also the clinically used dose interval. Details on dosimetry have been published previously.29 The plates were incubated in a water vapor-saturated atmosphere containing 5% CO₂ at 37°C. After 4 weeks, the number of positive wells was counted using an inverted phase-contrast microscope. Wells with colonies consisting of at least 32 cells were considered positive.

Time-lapse video microscopy

Twenty-four hours after plating, prior to starting the experiments, the cells were fed with fresh medium, or a medium containing 5 nM paclitaxel was applied. The medium was equilibrated with 5% CO₂ at 37°C in an incubator for 10 min. The culture flask was then capped and transferred to a 37°C heated stage of an inverted microscope (Nikon, Diaphot, Nikon, Tokyo, Japan) and the video recording was started.

Cells were viewed using phase-contrast optics at 20× objective magnification coupled to a JVC 3CDD KY-F30 video camera (Victor, Tokyo, Japan). An edge of a representative group of cells at the 40–60% confluence culture was selected for the field to be analyzed, containing approximately 20–60 cells. The time-lapse video recording was performed so that 2 successive pictures were taken at 30 sec intervals (Panasonic AG-6720A). The video recorder and the microscope were coupled to a timer (Dept. of Pathology, University of Turku, Turku, Finland), which lit the microscope lamp for 5 sec in every 30 sec in synchrony with the recorder. The culture flask was shielded from ambient room light.

Filming was continued for 48 hr. Subsequently the film was viewed frame by frame on video monitor. The cumulative numbers of mitoses and apoptoses per field were counted at 12 hr intervals. Mitosis was considered to have ended with the appearance of cell division. An apoptotic cell death was recorded either when a flat cell condensed rapidly (interphase apoptosis) or an already condensed cell (mitotic apoptosis) died after violent cytoplasmic pulsation and blebbing.

Data analysis

PE was calculated using the formula −ln (number of negative wells/total number of wells)/number of cells plated per well. Fraction survival data as a function of the radiation dose with or without indicated paclitaxel dose were found to fit in the linear quadratic equation. A microcomputer program was used to fit data to S = exp[−(αD + βD²)]. The area under the curve (AUC) value, equivalent to mean inactivation dose (D̄), was obtained by numerical integration. AUC ratios (AUC for 2 fractions of radiation/AUC for single-dose radiation) and (AUC for 2 fractions of radiation + paclitaxel/AUC for single-dose radiation + paclitaxel) were used to determine the SLDR capacity and the effect of paclitaxel on SLDR. A 2-pair t-test was used to test the significant differences in cell lines presenting SLDR.

To describe the type of interaction of the 2 modalities, we have used the term additive of the sum of individual effects. The term supraadditive is used if the combined effect exceeds the sum of individual effects. Some authors use the term synergy, which we have interpreted here as supraadditivity.

RESULTS

Seven vulvar SCC cell lines were evaluated in our study. The plating efficiencies, passages used, intrinsic radiosensitivity expressed as mean inactivation dose (AUC) and sensitivity to paclitaxel are summarized in Table II. All vulvar SCC cell lines were inherently radioresistant in vitro and AUC values were 1.8–3.3 Gy. The differences in paclitaxel sensitivity between the cell lines were small: inhibitory concentration of 50% (IC₅₀) values varied between 0.6 and 1.7 nM and the mean value was 1.2 nM. The effect of concomitant paclitaxel and radiation given in a single dose was clearly additive, as we have also shown earlier with 4 of the cell lines.10 The paclitaxel sensitivity or intrinsic radiosensitivity did not affect the synergy; the effect of concomitant paclitaxel and radiation was of the same magnitude as the combined effects of the 2 modalities given separately in all of the cell lines tested.

Table II. Sensitivity to Radiation and Paclitaxel1

Cell line	Plating efficiency	Passages (no.)	Intrinsic radiosensitivity (Gy) ±SD	Sensitivity to paclitaxel (nM) ±SD
1 Intrinsic radiosensitivity is expressed as the area under the survival curve (AUC) equivalent to the mean inactivation dose (D̄), and sensitivity to paclitaxel in vitro is expressed as IC50 values.
UM-SCV-1A	0.17–0.69	19–39	2.2 ± 0.1	1.7 ± 0.3
UM-SCV-2	0.35–0.76	21–27	2.3 ± 0.2	0.8 ± 0.1
UM-SCV-4	0.22–0.72	15–33	1.9 ± 0.1	0.6 ± 0.1
UM-SCV-6	0.11–0.29	19–30	1.8 ± 0.2	1.2 ± 0.4
UM-SCV-7	0.46–0.76	19–33	2.5 ± 0.2	1.2 ± 0.1
UT-SCV-3	0.022–0.11	18–21	2.5 ± 0.2	1.5 ± 0.5
A431	0.17–0.42	12–46	3.3 ± 0.3	1.3 ± 0.3

Three of the 7 vulvar SCC cell lines, UM-SCV-1A, -2 and -4, exhibited statistically significant SLDR (2-pair t-test) after splitting the radiation dose (Table III). The p-values were 0.000033, 0.000011 and 0.0022, respectively. The SLDR noticed was present at all radiation doses used. In most of the remaining cell lines a trend of SLDR was seen, as the AUC ratios between 2 fractions and 1 fraction were over 1.0, but the differences were not statistically significant.

Table III. Concurrent Paclitaxel and Radiation1

Cell line	AUC ratio ± SD
	RAD(S) + TAX/RAD(S)	RAD(F)/RAD(S)	RAD(F) + TAX/RAD(F)	RAD(F) + TAX/RAD(S) + TAX
1 Effect of paclitaxel (TAX) in combination with single-dose radiation [(RAD(S)] on clonogenic survival of SCV cell lines measured as area under the curve (AUC) ratios along with the AUC ratios between fractionated radiation [RAD(F)] and single-dose radiation and the effect of paclitaxel on it.
UM-SCV-1A	0.74 ± 0.06	1.14 ± 0.08	0.79 ± 0.14	1.08 ± 0.10
UM-SCV-2	0.86 ± 0.06	1.19 ± 0.07	0.78 ± 0.07	1.10 ± 0.10
UM-SCV-4	0.93 ± 0.07	1.05 ± 0.03	0.95 ± 0.05	1.07 ± 0.03
UM-SCV-6	0.76 ± 0.13	0.99 ± 0.09	0.82 ± 0.22	1.08 ± 0.24
UM-SCV-7	0.72 ± 0.05	1.04 ± 0.12	0.72 ± 0.10	1.08 ± 0.15
UT-SCV-3	0.71 ± 0.20	1.05 ± 0.12	0.74 ± 0.17	1.13 ± 0.26
A431	0.82 ± 0.15	1.08 ± 0.12	0.88 ± 0.28	1.13 ± 0.33

In 5 of the vulvar SCC cell lines, concomitant paclitaxel and radiation did not influence the effect of fractionation. Thus, the interaction of paclitaxel/split-dose radiation treatment was clearly additive and no change in the type of interaction was found between different radiation doses. In the cell lines UM-SCV-1A and UM-SCV-2, which exhibited significant SLDR capacity after dose fractionation, the AUC ratios did not show any clear SLDR after combining paclitaxel with the split-dose radiation. The interaction of the 2 modalities remained additive in these 2 cell lines also. However, the change in the amount of SLDR after splitting the radiation dose can readily be seen in the survival curves of the UM-SCV-1A cell line (Fig. 1).

Figure 1. Effects of concurrent paclitaxel and radiation in 7 vulvar SCC cell lines given either in a single dose or in 2 fractions. Fitted radiation survival curves with and without paclitaxel are presented. The results are given as the average of the actual data points and the bars represent 1 SD. Solid circles, single-dose radiation; open circles, 2 fractions of radiation; solid squares, single-dose radiation with paclitaxel; open squares, fractionated radiation dose with paclitaxel.

Time-lapse videomicroscopy was used to analyze the morphologic changes induced by 5 nM paclitaxel in 2 of the vulvar SCC cell lines, UM-SCV-1A and -7, which were grown in nonconfluent monolayers during the experiment. The results of the time-lapse videomicroscopy are shown in Table IV. As expected on the basis of our previous studies, the control cultures of both cell lines exhibited frequent mitosis (104% and 120%, respectively, of the initial cell number at 48 hr).30 Apoptoses were also seen in the control cultures; they represented 23% and 35%, respectively, of the initial cell number at 48 hr. In the cultures treated with 5 nM paclitaxel, the apoptotic frequency was moderately increased, to 39% and 58% of the initial cell number, respectively. However, the mitotic frequency was significantly reduced, to 0% and 27% of the initial cell number at 48 hr, respectively.

Table IV. Time-Lapse Videomicroscopy1

Cell line	Mitoses/apoptoses
	12 hr	24 hr	36 hr	48 hr
1 The mitotic and apoptotic frequency of the 2 vulvar cell lines (UM-SCV-1A and -7) after 5 nM of paclitaxel. The values used are cumulative and ratios to initial cell number; the numbers of events at different time points are given in parentheses.
UM-SCV-1A
Control (n = 26)	0.38/0.04 (10/1)	0.65/0.15 (7/3)	0.73/0.19 (2/1)	1.04/0.23 (8/1)
Paclitaxel (n = 28)	0/0.11 (0/3)	0/0.14 (0/1)	0/0.29 (0/4)	0/0.39 (0/3)
UM-SCV-7
Control (n = 20)	0.2/0.05 (4/1)	0.4/0.15 (4/2)	0.6/0.2 (4/1)	1.2/0.35 (12/3)
Paclitaxel (n = 60)	0.07/0.15 (4/9)	0.1/0.33 (2/11)	0.2/0.47 (6/8)	0.27/0.58 (4/7)

DISCUSSION

The effect of paclitaxel and fractionated radiation on 7 vulvar SCC cell lines was investigated in the current study. We had previously found an additive effect in 4 of these cell lines with paclitaxel and radiation given in a single dose.10 We wanted to approach clinical conditions by dividing the radiation dose into 2 equal fractions given at 24 hr intervals. This was done to determine the SLDR capacity of the cell lines and to evaluate whether paclitaxel could enhance the growth-inhibitory effect of radiation, for example, by inhibiting the repair of SLD.

We have previously tested the intrinsic radiosensitivity of these cell lines and the cell lines appeared to be relatively resistant as a group; AUCs ranged between 1.6 and 3.3 Gy.18, 19 Current results show modest SLDRs in these 7 vulvar SCC cell lines tested with a clonogenic method. Three of the cell lines, UM-SCV-1A, –2 and -4, exhibited repair of the SLD, the ratios between 1 and 2 fractions of radiation being 1.14, 1.19 and 1.05, respectively. Four of the 7 cell lines only exhibited a trend of repairing SLD. Our findings indicating limited ability to repair SLD in radioresistant vulvar SCC cell lines agree with recent investigations reporting that radiosensitive cells have a high SLDR capacity.28, 31–34 On the other hand, our findings are not in agreement with some previous suggestions that SLDR capacity and radioresistance had a direct correlation.35, 36 However, even if the SLDR capacity was low in vitro, there are other means by which the cytotoxic effect of the 2 modalities could be potentiated. The clear additive effect is favorable and could extend tumor control in clinical use.

Few preclinical reports of paclitaxel combined with fractionated irradiation are available. The number of fractions given varies between 2 and 5 and both single-dose and daily doses of paclitaxel are combined with radiation in these studies.37–40 With clinically relevant low paclitaxel doses combined with fractionated irradiation, a clear additive effect could be reached. Milas et al.41 showed that the radioresponse could be enhanced by tumor cell reoxygenation allowed by paclitaxel-induced apoptosis. This kind of enhancement would also be favorable in response to concurrent paclitaxel and fractionated radiation. In the current study the abolishment of the significant SLDR effect after adding paclitaxel to the treatment may indicate a true paclitaxel effect in 2 of the cell lines. Whether the effects of fractionated radiation could be potentiated through inhibition of SLDR by paclitaxel is of importance and should be investigated further with cell lines having a higher SLDR capacity. The additive effect achieved by combining paclitaxel with fractionated radiation in the radioresistant vulvar cell lines provides a new option against advanced vulvar cancer, which needs to be tested in clinical use.

It has been established that paclitaxel induces morphologic and biochemical changes consistent with apoptosis.42 Previously we demonstrated in laryngeal SCC cell lines that 10 nM paclitaxel induced a premitotic arrest after which the cells died by apoptosis.30 In our study time-lapse video microscopy was used to analyze the morphologic changes induced by 5 nM paclitaxel in 2 of the vulvar SCC cell lines studied. A dose of 5 nM paclitaxel induced moderately increased apoptosis compared with controls, but there was a clear decrease in mitotic frequency during 48 hr (Table IV).

Whether the enhanced radioresponse during combined use of paclitaxel and radiation is the consequence of cell accumulation in the radiosensitive G₂/M phase of the cell cycle has not been clarified. It has been suggested that the mitotic block is a critical determinant of paclitaxel-induced cell death,43 but Milross et al.44 found the antitumor effect of paclitaxel to correlate with paclitaxel-induced and baseline apoptosis, not with mitotic arrest, although mitotic arrest was apparent in all tumor types they reported, to varying degrees. In the current study 5 cell lines were also tested with flow cytometry after 24 and 48 hr of exposure to paclitaxel. A concentration of 1 nM, corresponding to those used in the clonogenic assay, was used. No accumulation of cells in the G₂/M phase was evident either after 24 or 48 hr of exposure (data not shown). There are reports suggesting, that supraadditive interaction could be seen without mitotic arrest;4, 5 furthermore, even if mitotic arrest occurs, the enhancement in radioresponse is not necessarily the result of this.6, 9 When malignant tumors are exposed to paclitaxel followed by radiation, the resulting growth inhibition or delay is achieved by several mechanisms on both cellular and molecular levels. When paclitaxel is combined with radiation, the growth inhibition is expected to be potentiated at least by paclitaxel-induced inhibition of sublethal repair and repopulation, as well as the apoptosis-enabled reoxygenation when in vivo models are used.

Much research is now focused on the influence of paclitaxel on the growth-regulating genes and the disturbances paclitaxel causes in signal transduction pathways along with the antiangiogenetic and antimetastatic activities and even the immunologic responses. During the last decade, ionizing radiation also has been shown to target the plasma membrane, where it may initiate multiple signal transduction pathways, leading to different biologic responses, including programmed cell death. Intense investigations in the near future into paclitaxel-induced apoptosis are likely to shed light on probable interactions between these 2 cytotoxic agents considering the intracellular pathways.

The radiosensitizing effect of paclitaxel does not seem to depend on the intrinsic radiosensitivity. Two rat yolk sac tumor cell lines with different radiosensitivities were found to be equally radiosensitized by paclitaxel.45 Geard and Jones7 suggested that radioresistant cells provide greater scope for discerning a response to the combined therapy; in our panel of inherently radioresistant vulvar SCC cell lines no variation was evident in the effect of concurrent paclitaxel and radiation on account of intrinsic radiosensitivity.

Despite the somewhat modest findings in the preclinical studies with the combination of paclitaxel and radiation, it has already become a well-proved treatment modality in several cancers, including nonsmall cell lung cancer.15 The treatment regimens are often complemented with the long-established platinum compounds. The promising results in these trials support the suggestion that the clinical response is more complex (involving several other cytotoxic mechanisms) than has been detected in preclinical experiments so far.

In conclusion, our results in the present study and our previous findings for single-dose radiation with paclitaxel, along with the promising results in clinical use in other malignancies including gynecologic tumors, encourage the planning of clinical testing with concurrent paclitaxel and radiation in vulvar cancer.

Acknowledgements

We express our gratitude to Mrs. M. Potila for her excellent technical assistance in performing the experiments. Paclitaxel (Taxol®) was provided by Bristol-Myers Squibb, Finland.